BSS Function Description

5/13/2018 BSS Function Description - slidepdf.com

http://slidepdf.com/reader/full/bss-function-description 1/158

Feature DescriptionM900/M1800 Base Station Subsystem Table of Contents

Huawei Technologies Proprietary

i

Table of Contents

Chapter 2 BSS Functions .............................................................................................................2-1

2.1 Basic Functions..................................................................................................................2-1

2.1.1 Overview .................................................................................................................2-1

2.1.2 Channel...................................................................................................................2-2

2.1.3 System Information .................................................................................................2-9

2.1.4 Idle Mode Behavior ...............................................................................................2-15

2.1.5 PLMN Selection.....................................................................................................2-18

2.1.6 Cell Selection and Reselection .............................................................................2-19

2.1.7 Location updating..................................................................................................2-24

2.1.8 Access...................................................................................................................2-32

2.1.9 Paging ...................................................................................................................2-33

2.1.10 Immediate assignment........................................................................................2-35

2.1.11 Assignment..........................................................................................................2-44

2.1.12 Authentication......................................................................................................2-45

2.1.13 Ciphering.............................................................................................................2-48

2.1.14 DTX .....................................................................................................................2-52

2.1.15 Frequency hopping .............................................................................................2-55

2.2 Extended Functions .........................................................................................................2-60 2.2.1 Handover...............................................................................................................2-60

2.2.2 Power Control........................................................................................................2-74

2.2.3 Extended Cell........................................................................................................2-86

2.2.4 IUO........................................................................................................................2-89

2.2.5 "HW-IUO Property"Satellite Transfer.................................................................... 2-95

2.2.6 Diversity Receiving................................................................................................2-97

2.2.7 Aggressive Frequency Reuse Pattern .................................................................. 2-99

2.2.8 Multiband Network ..............................................................................................2-104

2.2.9 Carrier Mutual-assistance ...................................................................................2-116

2.2.10 Cell Broadcast...................................................................................................2-119

2.2.11 Radio Channel Allocation..................................................................................2-121

2.2.12 Half Rate ...........................................................................................................2-125

2.2.13 E1 Ring Topology..............................................................................................2-127

2.2.14 GSM900/GSM1800 Co-cell...............................................................................2-129

2.2.15 Multi-MNC .........................................................................................................2-131

2.2.16 E-GSM/R-GSM..................................................................................................2-135

2.3 GPRS Function..............................................................................................................2-137

2.3.1 Supported Packet System Information ............................................................... 2-137

2.3.2 Supported GPRS MS Modes ..............................................................................2-141



Feature DescriptionM900/M1800 Base Station Subsystem Table of Contents


ii

2.3.3 Supported RLC Modes........................................................................................2-143

2.3.4 Supported Channel Coding Scheme ..................................................................2-144

2.3.5 Supported Network Control Modes.....................................................................2-148

2.3.6 Supported Network Operation Mode .................................................................. 2-148 2.3.7 Supported QoS....................................................................................................2-150

2.3.8 Supported Assignment........................................................................................2-150

2.3.9 Supported Paging ...............................................................................................2-151

2.3.10 Timing Advance ................................................................................................2-152

2.3.11 Measurement Report ........................................................................................2-153

2.3.12 Supported Flow Control ....................................................................................2-153

2.3.13 Supported Dynamic Handover between TCH and PDCH ................................ 2-155

2.3.14 Supported Packet Access Function..................................................................2-155



Feature DescriptionM900/M1800 Base Station Subsystem Chapter 2 BSS Functions


2-1

Chapter 2 BSS Functions

BSS is a bridge between MS and NSS, which performs mainly the management of

radio links and conversion of radio links and wire links. It is responsible for the

communication of MS. BSS system functions can be divided into basic functions,

extended functions and GPRS functions.

2.1 Basic Functions

2.1.1 Overview

Figure 2-1 illustrates the GSM Protocol.

CM

MM

RR

LAPDm

Sign.Layer1

L3

L2

L1

BTSM

MS

Um

SCCP

MTP

BTSM

RR BSSMAP

Abis

BTS

B

MSC

A

BSC

Sign.Layer1

Sign.Layer1

Sign.Layer1

RRLAPDLAPDm LAPD

CM

MM

BSSMAP

SCCP

MTP

MS: Mobile Station BTS: Base Transceiver StationBSC: Base Station Controller RR: Radio Resource ManagementMSC: Mobile services Switching Centre, Mobile Switching CentreMTP: Message Transfer Part (MTP) SCCP: Signaling Connection Control PartLAPD: Link Access Procedure on the D channel MM: Mobility ManagementLAPDm: Link Access Procedure on the Dm channel CM: Connection ManagementBSSMAP: Base Station Subsystem Management Application Part

BTSM: Base Transceiver Station Site Management

Figure 2-1 GSM protocol stack

According to GSM 04.07, the functions of BSS on layer 3 and related sub-layers on the

radio interface (Um) are classified into:

1) RR: Radio Resource Management

2) MM: Mobility Management

3) CM: Communication Management

Where the functions on the MM and CM sub-layers are supported by the DTAP

between A- and Um interfaces. The functions of RR sub-layer that include the





2-2

maintenance and release of radio resources are mainly carried out by BSS. There are

corresponding communication management protocol for A interface and Abis interface

to realize the air interface between GSM network and MS. The other functions of BSS

are also essential for establishing communication between the GSM network and MS.

The functions (RR) that BSS involves are mainly as follows:

Radio channel management

Channel coding/decoding

Transcoding & Rate Adaptation

Full-rate & half-rate coding of speech and enhanced full-rate coding

Encryption/Decryption

Frequency hopping

Antenna Diversity

RF Power control and handover management

2.1.2 Channel

I. Types of Radio Channels

According to GSM/GPRS specifications, the radio channels fall into two major

categories, which are Traffic Channel and Control Channel. A traffic channel s further

divided into Speech Traffic Channel, Circuit Data Traffic Channel and Packet Data

Traffic Channel, while the Control Channel is subdivided into Broadcast Channel,

Common Control Channel and Dedicated Control Channel.

Logical

channel

CCH

CCCH

DCCHTCH

SDCCH ACCH

BCCH

Downlink Uplink

FCCH

(BCCH1)

SCH

(BCCH2)

BCCH

(BCCH3)PCH AGCH RACH SACCH FACCH

Downlink

Downlink/Uplink

Figure 2-2 GSM/GPRS channel classification





2-3

Figure 2-2 illustrates the logical channels. Below is the introduction.

II. Traffic Channel

1) Speech traffic channels

In the latest GSM 05.02, the speech traffic channels are divided into:

TCH/FS: full rate traffic channel for speech.

TCH/HS: half rate traffic channel for speech.

TCH/EFS: enhanced full rate traffic channel for speech.

TCH/AFS: adaptive full rate traffic channel for speech.

TCH/AHS: adaptive half rate traffic channel for speech.

Huawei BSS currently supports three types of traffic channels for speech: TCH/FS,

TCH/HS and TCH/EFS.

2) Circuit data traffic channel

In the most updated GSM 05. 02, the circuit data traffic channels are divided into:

TCH/F9.6: full rate traffic channel for 9.6 kbit/s user data.


TCH/H4.8: half rate traffic channel for 4.8 kbit/s user data.

TCH/H2.4: half rate traffic channel for 2.4 kbit/s user data.


TCH/F14.4: full rate traffic channel for 14. 4 kbit/s user data.

E-TCH/F28.8: enhanced circuit switched full rate traffic channel for 28.8 kbit/s user

data.


data.


data.

Huawei BSS currently supportsTCH/F9.6, TCH/F4.8 and TCH/F2.4.

3) Packet Data Traffic Channel

There are two rates for the PDTCH:

PDTCH: full-rate PDTCH. With GMSK modulation it can carry packet data whose

momentary rates are 0~22.8 kbit/s, while PDTCH with an 8PSK modulation system can

carry packet data whose momentary rates are 0~69.6 kbit/s.

PDTCH is a one-way channel and categorized by the direction as:

PDTCH/D: downlink PDTCH, for MS terminated packet transmission.

PDTCH/U: uplink PDTCH, for MS originated packet transmission.





2-4

III. Broadcast Channel (BCH)

BCH is used to transmit broadcast messages to the MS in down link direction. It

includes the following logical channels:

1) FCCH (Frequency Correction Channel): This channel is responsible for

transferring the frequency correction signals to the MS so that the MS can be

adjusted to the corresponding frequency.

2) SCH (Synchronization Channel): This channel is responsible for transmission of

the frame synchronization number (TDMA frame number) and the Base Station

Identity Code (BSIC) to the MS.

3) BCCH (Broadcast Control Channel): This channel transmits the information

common to all cells, such as Location Area Identity (LAI), cell maximum allowable

output power, BCCH carrier frequency of the adjacent cells, and packet service

system parameters.

4) PBCCH (Packet Broadcast Control Channel): This channel transfers the

messages related to packet services.

5) Cell Broadcast Channel (CBCH): This channel is used for the cell broadcast

short message services. It uses the same physical channels as SDCCH.

The channels introduced above are downlink channels.

IV. Common Control Channel (CCCH)

CCCH are classified into the following four channels:

1) Paging Channel (PCH): Downlink channel. MS tunes to and receives the

information from this channel to check for any call from MSC at regular intervals.

2) Random Access Channel (RACH): Uplink channel, through which an MS

accesses the network and requests for allocating SDCCH.

3) Access Grant Channel (AGCH): Through which the network notifies the MS about

the allocation of the dedicated channel.

4) NCH (Notification Channel): Downlink channel used for Voice Group Call Service

(VGCS) and Voice Broadcast Service (VBS).

V. Packet Common Control Channel (PCCCH)

PCCH includes the following four channels:

1) PPCH (Packet Paging Channel): Downlink packet paging channel. MS tunes to

the PPCH channel at a regular interval to check if there is any call from SGSN.

2) PRACH (Packet Random Access Channel): Uplink packet random access

channel. MS requests to access the network via the PRACH channel.

3) PAGCH (Packet Access Grant Channel): Downlink channel. The network

notifies the MS of the allocation of the packet data traffic channels via the PAGCH

channel.





2-5

4) PNCH (Packet Notification Channel): Downlink channel, designed for

point-to-multipoint multicast call.

Huawei BSS supports PPCH, PRACH and PAGCH.

VI. Dedicated Control Channel (DCCH)

DCCH consists of the following channels:

1) SACCH (Slow Associated Control Channel): Associated with the SDCCH or

TCH. This channel is designed for MS to send received signal quality and signal

intensity of adjacent BTSs to the network, and meanwhile receives the system

information including transmission power, power adjustment and timing advance.

SACCH can be further divided into:

SACCH/TF: SACCH associated with TCH/F. SACCH/TH: SACCH associated with TCH/H.

SACCH/C8: SACCH associated with SDCCH/8.

SACCH/C4: SACCH associated with SDCCH/4.

SACCH/M: SACCH associated with TCH/F for multi-TS configuration.

2) FACCH (Fast Associated Control Channel): FACCH implements transmission

by occupying a part on TCH, mostly for transmitting handover command.

FACCH can be further divided into:

FACCH/F: FACCH associated with TCH/F;

FACCH/H: FACCH associated with TCH/H.3) SDCCH (Standalone Dedicated Control Channel): it serves to transmit the

signaling such as short message information, location updating information, etc.

between the MS and the network, prior to the call setup.

SDCCH/8SDCCH/8

SDCCH/4SDCCH/4

VII. Packet Dedicated Control Channel

1) PACCH (Packet Associated Control Channel): Downlink channel serving to

transmit the signaling, including response messages and power control messages,

to the MS. PACCH can also transmit the resources allocation and re-allocation

messages. PACCH shares the resource with the PDTCH currently allocated to MS.

When MS is in transmission mode, SGSN can page the MS via PACCH to initiate

CS service.

2) PTCCH/U (Packet Timing Advance Control Channel Uplink): PTCCH/U sends

the timing advance by way of random access burst when the MS operates in a

transmission mode.

3) PTCCH/D (Packet Timing Advance Control Channel Downlink): PTCCH/D is

designed to send transmission timing advance to several MSs. One PTCCH/D

corresponds to several PTCCH/Us.





2-6

VIII. Radio channel management

Radio channel management involves the management of diverse radio channels in the

GSM/GPRS. This process occurs in the phase of connection setup, maintenance,

modification and release.

IX. Radio channel combination

As per the logical channel types as listed above, a user can configure the following

channel combinations in the M900/M1800 BSS.

TCH/F+FACCH/F+SACCH/TF

SDCCH/8+SACCH/C8

FCCH+SCCH+BCCH+CCCH

FCCH+SCCH+BCCH+CCCH+SDCCH/4+SACCH/C4 BCCH+CCCH

BCCH+CBCH

SDCCH+CBCH

PBCCH+PCCCH+PDTCH+PACCH+PTCCH

PCCCH+PDTCH+PACCH+PTCCH

PDTCH+PACCH+PTCCH

X. Traffic channel management

BSS is in charge of all the configured traffic channels. When a call is established, MSC

sends the channel type, channel code and other parameters regarding the call to BSS,

which chooses a traffic channel based on the messages. BSS also assumes the task

for the measurement and release of these traffic channels.

XI. Dedicated control channel management

BSS manages all the available dedicated control channels. After MS has sends a

random access request via RACH or PRACH, BSS will allocate a DCCH for the MS.

Besides, BSS is also responsible for monitoring and releasing the link of DCCH.

XII. Broadcast channel and common control channel management

The management of the available broadcast channels and common control channels

by the BSS involves DRX management, paging message dispatching, AGCH and

PAGCH control, RACH and PRACH control, and BCCH message broadcast.

XIII. Terrestrial channel management

The management of terrestrial channels between BSS and MSC is to keep the

terrestrial circuit states at BSS and MSC consistent so that an idle circuit can be





2-7

available when MSC makes a call (“assign circuit”) and when MS performs handover

(“assign terrestrial circuit”). This is to ensure the success for the call and the handover.

Procedures included in the A-interface circuit resource management are Circuit

Block/Unblock, Circuit Group Block/Unblock, Unequipped Circuit, and Reset Circuit.

General principles of the circuit control includes:

Circuit management message is normally initiated by BSC. While resetting circuit

can be initiated either by MSC or BSC,

MSC can only block or unblock its circuits without affecting the circuits at the BSS

side.

The BSS can not change the circuit state that has been changed at the local end of

the MSC. For circuits blocked on the maintenance console at MSC side, the BSS

has no authority to unblock or reset the circuit.

XIV. Channel Coding & Decoding

The messages are encoded/decoded before being transmitted on the radio channel to

avoid radio channel interference. There are various coding and interleaving methods

for different logical channels (speech, data and signaling). For a detailed description of

the coding methods for various channels, please refer to the specifications GSM 05.

03.

XV. Transcoding & Rate Adaptation

Transcoding (TC) and Rate Adaptation provides an interface between the standard 64

kbit/s transmission at NSS side and the lower rate transmission at BSS side.

The conventional voice-coding mode is PCM with a rate of 64 kbit/s. It is widely applied

to PSTN. Pulse Code Modulation (PCM) is used for normal speech in PLMN, at a rate

of 64 kbit/s whereas in GSM, RPE-LTP or CELP coding with much lower rate (16 kbit/s)

is used due to the limitation of radio channel resources. To further improve the voice

quality, EFR (Enhanced Full Rate) is introduced. To implement EFR, newly designed

algorithms are used but it does not affect the coding rate on the Um interface. When

adopting EFR, the compression algorithm for the MS and Transcoder & Rate Adaptor Unit (TRAU) must be modified.

Generally, 3.6 kbit/s and 6 kbit/s data rates on the Um interface are arranged for the 8

kbit/s or 16 kbit/s channel (for transmission either on the full-rate channel or the

half-rate channel), while the 12 kbit/s rate is for the 16 kbit/s channel.

If a PSTN subscriber wants to call an MS, rate adaptation must be performed for the

voice. The TRAU is introduced to complete this function. When the BTS and the TRAU

are physically detached, these conversions will be especially important. A detailed

description of the conversions on the interfaces is given in the related GSM

specifications.





2-8

Since the rate of each channel of existing terrestrial lines is 64 kbit/s, it is a waste if one

channel is used to carry one 16 kbit/s GSM channel. To save terrestrial line resources,

sub-multiplexer (SMUX) is used between MSC and BSC to multiplex 4 % 16 kbit/s

channels to transmit four speech channels over one terrestrial channel.

In general, TRAU and SMUX are integrated in one unit called TCSM, i. e., it handles

both rate conversion and multiplexing.

Table 2-1 introduces the full-rate coding/decoding process and enhanced full-rate

coding/decoding process.

Table 2-1 Voice coding comparison

FR (Full Rate) EFR (Enhanced Full Rate)

Algorithm RPE-LTP algorithm (regular impulseexcitation-long term prediction) ACELP algorithm (arithmetic code bookexcitation linear prediction)

CodingProcess

TRAU converts the voice signal receivedfrom MSC into frames in the format of 20ms/fr. A frame of voice data contains 160PCM sampling points, making up 1280 bit.The output parameters after encoding are260 bit, making up the 320 bit TRAU frametogether with the synchronous header andcontrol parameter.

TRAU converts the voice signal received fromMSC into frames in the format of 20 ms/fr. Aframe of voice data contains 160 PCMsampling points, making up 1280 bit. Theoutput parameters after encoding are 244 bit,making up the 320 bit TRAU frame together with the synchronous header and controlparameter.

DecodingProcess

Decoding is a reverse process of coding. After TRAU receives the TRAU frames sent

from the BSC, it restores them into speechdata by applying decoding algorithm beforesending them to MSC.

Decoding is a reverse process of coding. After TRAU receives the TRAU frames sent from the

BSC, it restores them into speech data byapplying decoding algorithm before sendingthem to MSC.

In the occasion of MS-MS session, the TRAU coding / encoding can be omitted. As the

coding / encoding process will degrade the voice quality, it is possible to improve the

voice quality by removing TRAU coding/decoding with Tandem Free Operation (TFO).

TFO is implemented by FTC via in-band signaling to reduce the primary

coding/decoding during MS-MS session and improve the voice quality.

To set up TFO status, the following should be realized: Both parties of the session

should subscribe to the same service (i.e. both to FR or EFR service). The FTCs seized

by the two MSs should support TFO function. There should be no other equipment that

is capable of changing the PCM signal on the PCM link between the FTCs of the MSs,

i.e., it should be a direct link, because TFO message and frame are transmitted with the

low bit of the PCM sampling value. If these conditions are not satisfied, FTC will

perform the normal coding/decoding.

TFO features:

Realized in the occasion of MS-MS session;





2-9

TFO can improve the voice quality of both FR and EFR, especially the former with the

MOS can be improved by 0.5 points (totally 5 points).

2.1.3 System Information

I. Overview

System information contains the major wireless network parameter on the air interface,

including network identifier parameter, cell selection parameter, system control

parameter and network function parameter. By receiving system information, MS can

be properly accessed and perform network selection so that it can make full use of the

services and cooperate with network.

There are two modes for the transmission of system information: broadcast messageand channel associated message.

In idle mode, MS communicates with the network via the broadcasting of system

information. The network sends system information to MS so that MS knows its current

position and the service type available. Some parameters can also control the cell

reselection of MS.

When MS is establishing calls, the communication between network equipment is

realized with the channel associated system information. Network equipment sends

some contents in the channel-associated message to MS so as to control the behaviors

such as transmission, power control and handover of MS.

The broadcast system information is closely related to the channel-associated

message. The content in the broadcast system information can overlap with that in the

channel associated message. While the content in the channel associated message

can be inconsistent with that in the broadcast system information, because the channel

associated message has the effect on only one MS, while the broadcast system

information affects all MSs in idle mode.

II. Types and content of system information

There are totally 13 types: 1, 2, 2bis, 2ter, 3, 4, 5, 5bis, 5ter, 6, 7, 8 and 9. Among them,

1, 2, 2bis, 2ter, 3, 4, 7, 8 and 9 are broadcast information transmitted via BCCH under

idle mode; 5, 5bis, 5ter and 6 are channel associated information transmitted via

SACCH in active mode.

Type 1: Cell channel description + RACH control information (optional)

Cell channel description: all frequencies used by this cell, including BCCH frequencies

and FH frequency to provide the frequency reference for MS Frequency Hopping (FH).





2-10

RACH control information: parameters such as maximum times of retransmission

(MAX RETRANS), number of transmission timeslots (TX Integer), Cell Bar Access, bit

allowed for call reestablishment (RE), bit allowed for emergency call (EC) and access

restricted user level (AC). These parameters are used to control the behavior of MS inthe initial access.

Type 2: Adjacent cell BCCH frequency description + Network color code allowed +

RACH control information (mandatory)

Adjacent cell BCCH frequency description: the BCCH frequency used by the adjacent

cell.

Network color code allowed: NCC allowed for the MS test on the BCCH carrier in the

cell.

Type 2bis: Adjacent extended cell BCCH frequencies description + RACH controlinformation (optional)

Extended adjacent cell BCCH frequency description: the number of frequencies

described in the frequency allocation table in system information type 2 is limited,

therefore system information type 2bis contains the information of other frequencies in

BA1 which are in the same frequency segment as system information type 2.

RACH control information: contains the maximum times of parameter retransmission

(MAX RETRANS), number of retransmission timeslot (TX Integer), Cell Bar Access, bit

allowed for call reestablishment (RE), Restricted user level, bit allowed for emergency

call (EC) to control the MS behavior during initial access.

Type 2ter : Attached multi-frequency information + extended cell BCCH frequency

description 2 (optional)

Attached multi-frequency information: Number of the multi-frame measurement

needed.

Extended adjacent cell BCCH frequency description 2: describes the extended

frequency allocation table of the adjacent cell (part of BA1 table). The frequency

contained in this information is located at the different frequency segment as the

current cell. Therefore, only the multiband MS can read this information. The

single-band GSM 900 of GSM 1800 MS will skip this information.

Type 3:Cell ID + LAI + control channel description + cell option + cell selection

parameter + RACH control information (mandatory)

Cell ID: identifier of the current cell.

LAI: location area identifier of the current cell.

Control channel description: contains the MS attach/detach allowed indication (ATT,

Attach-Detach Allowed), number of blocks reserved for AGCH (BS AG BLKS RES),





2-11

common control channel configuration (CCCH CONF), number of 51 TDMA

multi-frames reserved for the same paging group in the paging information (BA PA

MFRMS) and the interval of periodic location update.

Cell option: includes the power control indication (PWRC), discontinuous transmission

(DTX) and radio link timeout value (Radio Link Timeout).

Cell selection parameter: includes the cell reselection hysteresis value, maximum Tx

power level allowed for MS access to the cell (MS TXPWR MAX CCH) and minimum

access level allowed for MS to access system (RXLEV Access MIN).





System information type 3 rest bytes: cell reselection parameter information and type 3

MS control information.

Type 4: LAI + cell selection parameter + RACH control information + CBCH description

+ CBCH dynamic allocation information (mandatory)

LAI: the location area identifier of the current cell.

Cell selection parameter: includes the cell reselection hysteresis value, maximum Tx

power level allowed for MS access to the cell (MS TXPWR MAX CCH) and minimum

access level allowed for MS to access system (RxLEV Access MIN).





CBCH description: includes the channel type and TDMA offset (which type of dedicated

channel combination), timeslot No. (TN), training sequence code (TSC), FH channel

indication (H), mobile allocation index offset (MAIO), FH serial No. (HSN) and absolute

RF channel No. (ARFCN).

CBCH mobile allocation information: the relation between the sequence of frequencies

used for FH and cell channel description.

System information types 4 rest bytes: cell reselection parameter.

Type 5: Adjacent cell BCCH frequency description (mandatory)

Adjacent cell BCCH frequency description: the BCCH frequency used by the adjacent

cell. Comparing with system information type 2, the difference is that MS can get the

frequencies described in system information type 5 in active mode, and report the

related information of the adjacent cell in the measurement report as the reference of





2-12

handover. Similarly, the GSM900 MS in Phase 1 recognizes only the adjacent cell

frequencies described in system information type 5 and ignore those contained in 5bis

and 5ter.

Type 5bis: Extended adjacent cell BCCH frequency description (optional)

Extended adjacent cell BCCH frequency description: the number of frequencies

described in the frequency allocation table in system information type 5 is limited,

therefore system information 5bis contains the information of other frequencies in BA2

which are in the same frequency segment as system information 5.

Type 5ter : Attached multi-frequency information + extended cell BCCH frequency

description 2 (optional)

Attached multi-frequency information: Number of the multi-frame measurement

needed.

Extended adjacent cell BCCH frequency description 2: describes the extended

frequency allocation table of the adjacent cell (part of BA2 table). The frequency

contained in this information is located at the different frequency segment as the

current cell. Therefore, only the multiband MS can read this information. The

single-band GSM 900 of GSM 1800 MS will skip this information.

Type 6: Cell ID + LAI + cell option (mandatory)

Cell ID: identifier of the current cell.

LAI: the location area identifier of the current cell.

Cell option: includes the power control indication (PWRC), discontinuous transmission

(DTX) and radio link timeout value (Radio Link Timeout).

Type 7: Cell reselection parameter

Cell reselection parameter: includes cell reselection indication (PI), Cell Bar Qualify

(CBQ), Cell Reselect Offset (CRO), Temporary Offset (TO) and Penalty Time (PT).

Type 8: Cell reselection parameter

Cell reselection parameter: includes cell reselection indication (PI), Cell Bar Qualify

(CBQ), Cell Reselect Offset (CRO), Temporary Offset (TO) and Penalty Time (PT).

Type 9: RACH control information + broadcast channel parameter


(MAX RETRANS), number of retransmission timeslot (Tx Integer), Cell Bar Access, bit



Broadcast channel parameter





2-13

III. Meaning and function of wireless network parameter

1) Network identification parameters

Network identification parameters include CGI and BSIC.

CGI consists of LAI and CI. LAI is composed of MCC, MNC and LAC. System

information type 3,4 and 6 include all or part of CGI information. MS decodes the

system information to get the CGI. MS decides whether to connect to the network in

this cell according to the MCC and MNC indicated by CGI. It is also used to check

whether the current location area has changed so as to initialize the location updating

process.

MCC, consisting of three decimal digits, is allocated worldwide in unified way. MNC,

consisting of two decimal digits, is allocated by the country in unified way. LAC and CI,

both consisting of 2 bytes, are arranged by GSM carrier in unified way. Note that the

value range of CI is 0X0001~0XFFFE, while 0X0000 and 0XFFFF are reserved.

BSIC identifies the local color code of each BTS in the GSM system. In GSM system,

frequencies are multiplexed to different extents according to the different requirements

in network plan. MS differentiates two cells' same frequency with their BSICs.

Therefore, it is necessary to guarantee the uniqueness of BSICs of the cells using the

same BCCH carrier frequency.

BSIC is transmitted on the SCH of each cell. It consists of NCC (3 bits) and BCC (3 bits).

Note that the TSC described in system information type 4 is the BCC of the current cell.

2) System control parameter

System control parameter is transmitted to MS with system information via air interface

by BTS. It serves to keep contact between MS and BTS. Besides, these parameters

have the direct effect on the service bearing and signaling flow of various part of system.

Therefore, reasonable setting of these parameters is important in maintaining of the

normal operation of GSM system.

IMSI attach and detach allowed (ATT) is used to notify MS whether the local cell

allows IMSI attach/detach process. It is transmitted in control channel description in the

system information type 3. ATT has 1 bit. "0" stands for IMSI attach/detach process not

allowed, and "1" stands for the process allowed.

CCCH CONF decides the integration mode of the CCCH in the cell. It is transmitted in

the control channel description in the system information type 3. CCCH CONF is a 3 bit

code. For details, see Table 2-2.





2-14

Table 2-2 CCCH code meaning

CCCHCONF

MeaningNumber of CCCH

information blocks

in BCCH multiframe

000CCCH uses a basic physical channel which is not shared withSDCCH.

9

001 CCCH uses a basic physical channel which is shared with SDCCH. 3

010CCCH uses two basic physical channels which are not shared withSDCCH.

18

100CCCH uses three basic physical channels which are not sharedwith SDCCH.

27

110CCCH uses four basic physical channels which are not shared withSDCCH.

36

Others Reserved

Note:

The CCCH CONF setting of a cell should be in line with the actual setting of the cell's CCCH. It is decided

by the traffic module of the cell.

BS AG BLKS RES is transmitted in the control channel of system information type 3. It

is used together with CCCH CONF to decide the number of information blocks in each

BCCH of the current cell. After setting CCCH CONF, BS AG BLKS RES will be used to

arrange the occupancy ratio between AGCH and PCH on CCCH. It is possible to adjust

this parameter to achieve the bearing balance between AGCH and PCH.

BS PA MFRAMS is transmitted in the control channel description in system information

type 3. It decides how many multiframes making up a cycle of a page sub-channel. This

parameter actually decides how many sub-channels the PCH of a cell will be deviled

into. BS PA MFRAMS is a 3 bit code. The value range is 0~7, respectively meaning that

the number of multi-frame of a paging group cycled on the PCH is 2~9.

Periodic location updating timer (T3212) decides the frequency of periodic location

updating. It is transmitted in the control channel description in system information type

3. It is an 8-bit code. The value range is 0~255, each unit of which is the duration of six

minutes, and 0 means no location updating.

Cell Channel Description, transmitted in system information type 1, describes the RF

channel No. of the local cell. It is used in frequency hopping. Note that the maximum

number of channels configured in cell channel description is 64.





2-15

Neighbor Cells Discretion, transmitted in system information type 2, 2bis, 2ter, 5, 5bis

and 5ter, describes the absolute channel No. of the BCCH TRX of the cell adjacent to

the current cell. Huawei BSS supports at most 32 adjacent cells.

Extension Indication, transmitted in system information type 2 and 5, indicates

whether there are still extended adjacent cells to be transmitted in system information

type 2bis and 5bis. It is a 1-bit code. "0" means that system information type 2 and 5

contains the complete BA table, and "1" means that type 2 and 5 contains part of BA

table.

BA Indication transmitted in system information type 2 and 5. It is a 1-bit code, used

for MS to select the data in BA 2 before or after modification. In another word, if the

adjacent cell relation of the current cell and the BA2 table is changed during a session,

the BA Indication in system information type 5 will be 1 instead of stead of 0. This

indicate that MS perform decoding in the adjacent cell indicated in the system

information type 5 again.

Multiband Reporting (MBR), transmitted in system information type 2ter and 5ter. It is

a 2-bit code, indicating MS to report adjacent cell information on multiple frequency

bands. It is applicable to multiband MS only.

2.1.4 Idle Mode Behavior

I. Overview

A powered on mobile station (MS) that does not have a dedicated channel allocated is

defined as being in idle mode. The purpose of the tasks performed in the idle mode is to

be able to access the system and be reached by the system from any location in the

network.

When a mobile is powered on, it immediately attempts to make contact with a GSM

Public Land Mobile Network (PLMN). The particular PLMN contacted may be selected

either automatically or manually. The MS will look for and select a suitable cell of the

chosen PLMN. It will then tune to the control channel of the cell to receive information

about the available services provided by the PLMN. This selecting is known as

“camping” on a cell. When an MS is in idle mode it will always try to camp on the best

cell according to a signal level based criterion.

The idle mode behavior is managed by the MS. It can be controlled by parameters

which the MS receives from the base station on the Broadcast Control Channel (BCCH).

All the main controlling parameters for idle mode behavior are transmitted on the BCCH

carrier in each cell. When the MS is powered on but neither making nor receiving any

calls (idle mode) there has to be a mechanism that always selects the best cell on

which to camp. Moreover, to be able to access the system from anywhere in the

network, regardless of where the MS was powered off, it has to be able to select a





2-16

specific GSM base station, tune to its frequency and listen to the system message

informations transmitted in that cell. It must also be able to register its current location

to the network so that the network knows where to route incoming calls. The PLMN

selection mechanism, the cell selection and reselection algorithms in addition to thelocation updating procedure are the core of the idle mode behavior. The purpose is to

always ensure that the mobile is camped on the cell where it has the highest probability

of successful communication.

II. Usage

1) High signal level when accessing the system

The MS will at all times try to obtain the highest possible signal level when accessing

the system. This is achieved by means of the idle mode cell selection and reselection

algorithms. These algorithms will enable the MS to choose the most suitable cell tocamp on, based on signal level. A cell is suitable if certain criteria are satisfied.

Camping on the most suitable cell provides the MS with a high probability of good

communication with the system.

The cell selection and reselection algorithms are governed by parameter settings.

Using these parameters an operator can, on a per cell basis, make a specific cell more

or less attractive to camp on for the MS. This makes it possible for the operator to

achieve similar behavior for MSs in idle mode as in active mode. Well-designed

parameter settings for cell selection and reselection in idle mode, will make the MS to

camp on the cell that would have been chosen if the MS had been in active mode.

2) Control of the paging load

In idle mode the MS will notify the network whenever it changes location area by the

location updating procedure. Thus, the network will be kept updated concerning which

location area the MS is presently in. When the system receives an incoming call it

knows in which location area it should page the MS, and does not need to page it

throughout the whole MSC service area. This reduces the load on the system. If the MS

does not respond to the first paging information, then the network can send a second

paging information.

The MS can also, periodically and when powered on or off, notify the network of its

present status by the location updating procedure. This prevents the network from

doing unnecessary paging of MSs that have been powered off or left the coverage area.

This would otherwise cause unnecessary load on the system.

3) Low idle mode power consumption

In idle mode, the MS only occasionally monitors the system information being

transmitted in the current cell or does measurements on neighboring cells to see if a

cell change should be initiated.





2-17

However, most of the time it will be in “sleep mode”. Hence, the power consumption

during idle mode will be low. This is also referred to as discontinuous reception (DRX).

III. Technical description

While the MS is in idle mode it will continuously make measurements on the

BCCH-carriers of serving and neighboring cells to decide on which cell to camp on. It

will also, if necessary, register its presence in the location area of the chosen cell by

performing a location updating.

The purpose of camping on a cell is threefold:

1) It enables the MS to receive system information from the PLMN

2) The MS can initiate a call by accessing the network on the Random Access

Channel (RACH) of the cell on which it is camped,

3) The PLMN will know the location area of the cell in which the MS is camped

(unless the MS has entered a limited service state) and can therefore page the MS

when an incoming call is received.

The idle mode task can be subdivided into four processes:

PLMN selection

Cell selection

Cell reselection

Location updating.

The relationship between these processes is illustrated in Figure 2-3.





2-18

Location Updating

Cell Reselectin

Automatic/Manual

Mode Selection

Indication to User

User Selection

of PLMN

Service

indication

to User

PLMNSelection

PLMN Available

PLMN Selection

Cell Selection

New

Location

AreaInitial Cell Selected

Periodic

Registration

Location UpdatingResponsesCell & Location

Area Changes

Figure 2-3 Overall idle mode processes

2.1.5 PLMN Selection

I. Overview

The MS will select a PLMN when it is powered on or upon recovery from a lack of

coverage. It will first try to select and register on the registered PLMN if one exists. If

registration on a PLMN is successful, the MS indicates this PLMN (the “registered

PLMN”) and is capable of making and receiving calls on it. If there is no registered

PLMN, or if the registered PLMN is unavailable, the MS will try to select another PLMN

either automatically or manually depending on its operating mode, The MS normallyoperates on its home PLMN. However, another PLMN may be selected if, for example,

the MS loses coverage. The MS will register on a PLMN if the MS finds a suitable cell to

camp on and if a location-updating request is accepted. Registration has to be

successful in order for the MS to be able to access that network.

However, it does not need to perform location updating if it is in the same location area

belonging to the same PLMN as it was before it entered the inactive state.

The MS can select and register on another PLMN of its home country than its home

PLMN if national roaming or international roaming is permitted. However, the MS will

then do periodical attempts to return to its home PLMN. This is controlled by a timer.





2-19

The interval between attempts is stored in the Subscriber Identity Module (SIM). Only

the service provider is able to set the timer value for return to home PLMN.

There are two modes for PLMN selection; automatic and manual. The automatic mode

utilizes a list of PLMNs in an order of priority whereas the manual mode leaves the

decision to the user and only indicates which PLMNs that are available.

II. Automatic mode

In automatic mode, the MS will select PLMN if available and allowable, in the following

order if no registered PLMN exists or is available:

Home PLMN

1) Each PLMN that has been stored in the Subscriber Identity

Module (SIM) in priority order

2) Other PLMNs with received signal level above -85 dBm in random order

3) All other PLMNs in order of decreasing signal level.

III. Manual mode

In manual mode, the MS will first try to select the registered PLMN or home PLMN (if no

registered PLMN exist). If this registration fails or if the user has initiated a PLMN

reselection the MS will indicate to the user all available PLMNs. The user can then

select a desired PLMN which causes the MS to initiate a registration on this PLMN. If

the selected PLMN is not allowable, an indication to the user to select another PLMN

will be made.

The user can at any time request the MS to initiate reselection and registration onto an

alternative available PLMN. This is done either using automatic or manual mode,

depending on the mode selected by the user.

2.1.6 Cell Selection and Reselection

I. Overview

The purpose of cell selection and reselection is to enable MS to find a most suitable cell

on which MS can reliably decipher the downlink data and maintain a high

communication rate on uplink (so as to realize various telecom services). Once MS has

selected a cell as its serving cell, its communication with the network becomes possible

on this cell.

MS will tune to the BCCH to receive the paging message and the system information

broadcast on BCCH and use the RACH to send access request after it has selected this

cell.





2-20

MS implements cell reselection according to the message in BA table in the system

broadcast information from the serving cell. There are two BA tables in GSM network.

One is transmitted in the system information via BCCH. It includes the BCCH carrier

used in a certain physical area for the MS in idle mode to implement cell selection andreselection. The other one is transmitted in the system information via SACCH. It is

used to indicate the MS in active mode about the BCCH carrier for handover

monitoring.

In active mode, MS obtains the information of adjacent cell BCCH frequency through

BA (BCCH). The process will not stop until MS receives the first BA (SACCH)

information.

II. Cell selection procedure

When MS is powered on and move from blind spot of coverage to the serving area, it

will search for all available frequencies in the PLMN and select the suitable cell to camp

on. This is the procedure of "cell selection".

Cell selection procedure in the case of no BCCH information in MS

MS first searches the 124 RF channels of GSM 900(if the MS is a multiband one, MS

searches 374 GSM 1800 RF channels), and compares the signal level on the channels

to calculate the average level. The entire measurement procedure lasts 3~5 s, during

which, at least five sampling points will be extracted from different RF channels.

After MS has tuned to the maximum carriers of the receiving level, it will first judge

which one is the BCCH carrier (by searching FCCH burst). If so, MS will attempt to

decode SCH to obtain the BCCH system broadcast information synchronous with this

carrier. If the MS can properly decode BCCH data, and make sure that this cell belongs

to the selected PLMN, parameter C1>0, and this cell has no access barring, MS can

camp on this cell. Otherwise, MS will keep tuning to the next highest carriers until it

reaches the available cell.

If no suitable cells are found after searching 30 RF channels with the highest level, MS

will monitor the level of all channels and search for the BCCH of C>0 and no access

barring. After finding this carrier, MS will camp on this cell without considering its PLMN

ID. In this occasion, only emergency call can be implemented.

Case 1: If the access level of the MS is barred at the cell, the cell selection algorithm will

not be affected, i. e., when the cell satisfies the criterion, MS will still camp on it.

Case 2: If the cell selected by MS belongs to PLMN, but access is barred (parameter

CBA is set as "bar") or algorithm C1<0, MS will use the BA table obtained from this cell

to search for these BCCH carriers.

Cell selection procedure in the case of BCCH information already stored in MS





2-21

If the BCCH carrier information has been stored in MS during the last powering off, MS

will first search the stored BCCH carrier. If MS can decode the BCCH data of the cell,

but cannot camp on it. MS will check the BA table of this cell. If no suitable cells found

after all BCCH carriers have been searched, the previous procedure will beimplemented.

C1 is the path loss criterion as the reference of cell selection and reselection. C1 of the

serving cell should be larger than 0. The formula is as follows:

( )( )0, _ _ _ _ _ 1 P CCH MAX TxPWRMS MAX MIN Access RxLEV RxLEV C −−−=

See Table 2-3 for formula explanation.

Table 2-3 Name of powers

Name Meaning (Unit d Bm)

RxLEV Average level MS received

RxLEV Access MIN Maximum receiving level allowed for MS to access

MS TxPWR MAX CCH Maximum transmitting power level allowed for MS to access the system

P Maximum output power of MS

C1 algorithm is used during cell selection procedure, as shown in Figure 2-4.

C1=15 C1=8

Cell1 Cell2

Figure 2-4 Cell selection

MS select the suitable cell to camp on according to the priority and C1. The selected

cell is the main serving cell Figure 2-4, MS will select Cell 1 as the main serving cell to

if the priorities are the same.

III. Cell reselection procedure

After MS has selected a serving cell, it will camp on this selected cell and continue the

monitoring on all BCCH carriers configured in the adjacent cell frequency configuration

table indicated in the BCCH system information of the serving cell if the conditions are

not changed greatly.





2-22

When monitoring these BCCH carriers, the measurement of their receiving level should

base on at least the average of 5 sampling points, and the number of measured

sampling points extracted from all BCCHs should be the same. The sampling points

allocated to each carrier should be as even as possible in each measurement period.The six strongest BCCH carriers should be refreshed at least once per minute.

To lower the power consumption of MS, MS should measure the receiving level of each

carrier in BA table when performing decoding page group. It is possible to obtain some

BCCH frequencies and sample values of receiving level on the BCCH frequency of the

serving cell during the appearance of MS page group.

The MS routine measurement program also includes the measurement of the BCCH

carrier of the current serving cell. MS should attempt to decode all system informations

broadcast on BCCH of the serving cell at least every 30 s. MS should implement

decoding of BCCH data block to the BCCH carriers of the six strongest non-serving

cells at least every 5 min. This data block contains the parameter concerning cell

reselection. After MS has found a new BCCH carrier as one of the strongest carriers, it

will decode the BCCH data of the new carrier within at least 30 sums. MS should check

the BSIC of one of the six strongest carriers within at least 30s to verify that the

monitored objective is the same cell. If BSIC is changed, MS will regard the carrier as a

new one, and decode the BCCH data again. During the process above, MS tries not to

interrupt the monitoring to PCH.

Under the following occasions, the procedure of cell reselection will be initiated. (If C2

algorithm has not been activated, C2 = C1).

MS finds that the C2 value of a cell (in the same location area as the serving cell) has

been larger than that of the serving cell for 5 seconds.

MS finds that the C1 value of a cell (not in the current location area) has been larger

than the sum of the C2 value of the serving cell and the cell selection hysteresis for five

seconds.

The current cell barred.

MS finds the downlink failure: the criterion of downlink signaling failure is based on thedownlink signaling failure counter DSC. If MS has selected a cell, DSC is set as [90/BS

PA MFRMS] round number. BS PA MFRMS is the number of multiframes of the 51

TDMA frame for the BTS transmission paging information for the MSs of the same

paging level. Therefore, when MS is decoding on the PCH, if succeeded, add 1 to DSC;

if failed, subtract 4 from DSC. When DSC = 0, there is downlink signaling failure.

The value of C1 has been smaller than 0 for 5 s.

During random access, MS fails to register at the retry after maximum retransmission.





2-23

Note that after MS reselection and camping on the cell, MS should decode all of the

BCCH data of the new cell to check whether the parameter concerning cell reselection

has changed. If it is changed, MS will decide whether this change satisfies the criterion

of cell reselection. If the criterion is satisfied, MS will camp on this cell. If MS finds thatLAI has changed, it will initialize location updating.

C2 algorithm is used in cell reselection, as shown in Figure 2-5

C2=4 C2=18

Cell1 Cell2

Figure 2-5 Cell Reselection

MS selects the cell to camp on according to the priority and C1 value. The camped-on

cell becomes the main serving cell. See Figure 2-5. With the same priority, MS will

select Cell 2 as the main serving cell if reselection hysteresis and the reselection time

are both satisfied.

IV. The impact of the network to the MS in idle mode

Network side is responsible for completing system informations broadcast and paging

task for idle MSs in downlink.

System information type 2~4 and the optional type 1, 2bis, 7 and 8 are broadcast

periodically from the network via BCCH. The MSs in idle mode decides whether and

how to access the network according to these information.

MS of GSM 900 supports the band of GSM 900 only. It regards the EXT IND bit

described in adjacent cell in system information type 2 as the standby bit. If theinformation sent from the multiband network is received, MS will regard that the

information unit in system information type 2 contains the complete BA table and will

ignore the system information type 2bis.

V. Definition of discontinuous receiving mode (DRX) and PCH

If MS in idle mode has selected its serving cell, it is ready to monitor the paging

information from this cell. To lower the power consumption of MS, the GSM

specification adopts the discontinuous receiving mechanism, i. e., each subscriber

(IMSI) corresponds to a dedicated paging group. Each group corresponds to a paging





2-24

sub-channel of the cell. MS recognizes its paging group and the corresponding paging

sub-channel according to the last three digits of the IMSI. MS in idle mode uses its own

paging sub-channel to receive the paging information (or to monitor the receiving level

of the BCCH carrier of the non-serving cell). MS ignores the information from other paging sub-channel or even shuts down the power of some hardware to lower its power

consumption during the broadcasting of other paging sub-channels. But MS must

measure the network information task periodically.

The number of the paging sub-channels can be calculated according to the

configuration type and BS AG BLKS RES (how many AGCH blocks for 51 multiframes),

BS PA MFRMS (how many 51 multiframes to make up a cycle of the paging

sub-channel).

Common Control Channel (CCCH) includes AGCH and PCH. It is used to transmit the

immediate assign information and paging information. CCCH can be bearded by a

physical channel or shared by multiple physical channels. CCCH can share the same

physical channel with SDCCH. The combination mode of CCCH is decided by the

parameter CCCH CONF. The configuration of CCCH CONF should be consistent with

that of CCCH. For the cell with one TRX, the recommended CCCH configuration is

sharing one physical channel with SDCCH (3 CCCH information blocks in this case).

For some location area with very heavy paging traffic, only one physical timeslot is

insufficient to transmit the paging information. Therefore, the GSM specification allows

configuring extra CCCHs on the TS0, TS2, TS4 and TS6 of the carrier.

2.1.7 Location updating

Location updating is an important task of Mobile Management (MM).

I. Location Area

To locate MS, each GSM PLMN domain is divided into locations areas covering one or

more cells. The location area of each MS is recorded by the network as the location

reference for paging this MS. With the introduction of the concept of location area, the

paging MS can be implemented with a location area instead of all cells controlled by

MSC, thus lowering the paging load. Each location area is assigned with a Location

Area Code (LAC), which is broadcast with the system information via BCCH.

The size of a location area has a great effect on the system. The design of location area

is very important in network planning. If the coverage of a location area is too small, the

location updating of MS will trigger frequently, which will increase the signaling flow of

system. On the other hand, if the coverage of a location area is too large, the load of

PCH and the signaling flow on Abis interface will increase since one single paging

information will be broadcast in all cells of this location area.





2-25

Therefore, optimization of location area is a very important task in network planning.

When designing the location areas, it is necessary to lessen the frequency of location

updating on the basis of no overweigh paging load, so as to avoid waste of network

resource.

II. Location updating

When MS roams from one location area to another, it is to be registered in the new

location area. In other words, once driven by certain needs or finding that the LAI stored

is different from that of the current cell, MS will notify the network to change the stored

location area. The procedure is called location updating.

If the MS in idle mode triggers cell reselection when moving within the same location

area, MS will not notify the network about this change although the serving cell has

changed. If the two cells before and after reselection are not in the same location area,

MS will notify the network about this change. This is "forced register".

According to the labels of location updating in the network, there are three types of

location updating: generic location updating (i.e. inter-location area location updating),

periodic location updating (T3212 timeout) and IMSI attach (MS powered on). Their

specific differences are whether only one VLR is involved in the location updating

process and whether IMSI is used in the process.

III. Generic location updating

Generic location updating is for the purpose of updating the actual MS's location

registered in the network. The information unit of type of location updating in "location

updating" should be indicated as generic location updating.

If the network indicates that the status of MS in VLR is unknown, the generic location

updating will also be initiated as a response to the request of MM connection setup.

1) Intra-VLR location updating

This is the simplest location updating process, in which, MS does not need to provide

its IMSI. It is implemented within the current VLR, and HLR will not be notified about the

process.

During the initialization process, the access cause indicated in the initialization

information contained in SABM frame sent from MS to the network is Location updating

Request. This information also contains MSTMSI and LAI noted as for generic location

updating. After receiving this information, MSC will send MAP Update Location Area to

VLR. VLR, after receiving this information, will implement the location updating. It will

update the location information of the MS and store the new LAI and then allocate a

new TMSI for MS if necessary (TMSI can also be absent in the TMSI reallocation

command. In this case, MS uses the former TMSI). After receiving TMSI Reallocation





2-26

Complete from MS, VLR sends Location updating Accept to MS, and then release the

channel to end the process.

IV. Inter-VLR location updating

If MS roams to a cell whose LAI is different from the current one, it will send the old LAI

and stored TMSI via MSC to VLR in the process of location updating If TMSI cannot be

identified, MS can also be identified with its IMSI. See Figure 2-6.

MS MSC VLR HLRPVLR

Location Update Request

MAP Update Location Area

MAP Update Location

MAP Insert Subscriber Data

MAP Insert Subscriber Data ACK

MAP Update Location ACK

MAP Update Location Area ACK

Location Update Accept

MAPCancel Location

MAPCancel Location ACK

A B D

D

Figure 2-6 Interfaces and process of inter-VLR location updating

1) Update with TMSI

If VLR finds that the TMSI is unknown after receiving MAP Update Location Area from

MSC, it will label the "VLR Location Information Acknowledge" as "Unacknowledged"

for the subsequent updating in HLR. If the subscriber has not registered in that VLR,

"HLR Location Information Acknowledge" will be labeled as "Unacknowledged". And

then, according to the address of the previous VLR (PVLR) indicated in TMSI and LAI,

VLR will send MAP Send Identification to PVLR to request for IMSI and authentication

parameter, and as a response PVLR will return the IMSI and authentication parameter

to the new VLR. If the new VLR fails to get the IMSI, it will then sends Identity Request

to MS to request for its IMSI. After receiving IMSI, VLR will send the information of

location updating of MS to HLR. This information contains the identification of MS and

other related information for HLR the query the data and set up the path. If the new





2-27

MSC/VLR has the normal service authority, HLR will store the new VLR No., and sends

MAP Cancel Location to HLR. After receiving this information, PVLR will delete all

information related to this MS, and sends MAP Cancel Location ACK to HLR. The new

VLR continues to handle the processes of authentication, ciphering and TMSIreallocation. When these processes are done, HLR sends MAP Insert Subscriber Data

to VLR to provide the subscriber information needed, including authentication

information. After receiving the response from VLR, HLR will send Location updating

Ark to that VLR.

2) Update with IMSI

If the identification of the subscriber is IMSI, VLR will check whether this subscriber is

unknown. If so, it will be labeled the "HLR Acknowledge" as "Unacknowledged", and

then initializes HLR updating. If the IMSI is a known one, VLR will check whether the

previous LAI provided in the information from MSC belongs to this VLR. If not, it willlabel "HLR Acknowledge" as "Unacknowledged", and then initialize HLR updating.

Authentication is needed in these two cases.

V. IMSI attach and detach process

IMSI attach and detach means to attach a binary mark to the subscriber record in

MSC/VLR. The former one is marked as access granted, and the latter one is marked

as access denied.

IMSI attach and detach is an option of system. If the cell where MS is powered on

supports this function, it will notify its power-on status to the network, i. e. sending the

information of "IMSI Attach" to notify the network about the change of its current status.

When the network receives this indication, it will note down the subscriber status in the

system data so as to initialize the paging process when there is an paging information

of this MS.

If MS finds that the stored LAI is the same as the current LAI when powered on, it will

initialized the process of IMSI attach. The process is almost the same as INTRA VLR

Location Updating. The only difference is that the type of location updating marked in

Location Updating Request is IMSI attach.

VI. Periodic location updating

Periodic location updating is used to periodically notify the network about the

accessibility of MS. MS sends Location Updating Request to the network, in which the

information unit of the type of location updating is periodic location updating.

In the following cases, network will lose contact with MS:

A powered on MS roams to the area beyond network coverage (blind spot). Since the

network is not notified about the current status of MS, it still considers the MS in the

status of IMSI attach.





2-28

1) When MS is transmitting "IMSI Detach", if there is interference to the radio uplink

path, the network may not be able to decode this information. This means that

system still regards this MS in the IMSI attach status.

2) In the case of MS power failure, MS cannot notify the network about its currentstatus, resulting in the loss of contact with the network. If the above cases happen

and the MS is paged, system will still sends the paging information to the location

area where the subscriber registered. This paging will sure end up with paging

timeout, and system resources are wasted.

To tackle this problem, the corresponding measure is taken in GSM system to make the

MS automatically reports its current location information to the network periodically. N

this way, the network can have the timely information of the current location status of

MS. This process is called periodic location updating. BSS sends the period of periodic

location updating (T3212) to all subscribers in the cell with system broadcast system

via the cell's BCCH, so that MS will automatically initialized location updating request to

the network when the timer times out. After cell selection or cell reselection, MS will

read T3212 from the system information of the serving cell, and then activate this timer

and store it in SIM. After that, whenever T3212 times out, MS will automatically initialize

location updating. At NSS side, the network will periodically query the subscribers

marked as IMSI attach in its VLR to mark those without any contact with it witting this

period as implicit power-off in order to avoid paging these MSs and wasting system

resources.

Periodic location updating is an important measure to keep the contact between the

network and MSs, therefore, the more frequent periodic location updating is, the better

overall performance of network can be achieved. However, frequent periodic location

updating has two drawbacks:

Increase of signaling flow which may lower the processing power of MSC/BSC/BTS

and the utilization of radio resources if the situation is serious;

Increase of MS power consumption which will shorten the standby time of the MSs

served by this system. Therefore, the setting of T3212 should be based on the actual

situation.

In the following cases, T3212 will be reset to 0:

When receiving "Location Updating Request" or "Location Updating Refuse",

Ciphering mode complete when receiving the first MM message, or MM connection

being established.

MS responds to its paging, and after which, it receives the first correct L3 message

(excluding RR message).

T3212 timed out.

MS deactivated (equipment powered off or SIM removed).





2-29

When T3212 times out, MS will initialize periodic location updating.

If T3212 times out when MS is in the status of "no available cell", or "service restricted"

or "searching for PLMN, MS will delay location updating will be delayed until these

status changed. If BCCH information indicates periodic location updating not applied,

this process will not be activated. T3212 timeout value is broadcast in the CCH

description in "system information type 3".

In the status of "no available cell", "service restricted" and "searching PLMN", T3212

cannot be changed.

MS, after cell reselection, may find that the T3212 of the new cell is different from the

previous one (sharing the same LAC) or the broadcast T3212 of the current cell is

manually changed). In this case, assumed that "t1" is the new T3212 timeout value and

"t" is the current value of T3212, the timer of MS will be restarted with the value of t modt1.

If MS is in the activated status, or it is necessary to change T3212 value, and the timer

is not running, then the new timer will be started with a random number whose value

range is 0~t1 ("t1" is new T3212 timeout value.

The signaling flow of periodic location updating is the same as that of generic location

updating.

VII. Generic location updating (specification)

MS initialize location updating

If there is no RR connection available when initializing location updating, MM sub-layer

of MS will request RR sub-layer to establish RR connection.

MS sends "Location Updating Request" to the network, and start T3210. The

information unit of location updating type in this message will indicate the type of this

location updating. On this occasion, the network can initialize the type querying

procedure (e. g. to get the ciphering capability of MS). If the network cannot obtain the

IMSI according to TMSI and LAI, the network can initialize the identification process.

After receiving "Location Updating Request" from MS, the network will initialize the

authentication process. If it is necessary to reallocate TMSI, the network will initiate the

process of ciphering mode setting.

1) Attempt counter

To restrict the frequency of location updating attempt, the attempt counter is

recommended in the specification. It is used to count the number of consecutive

unsuccessful location updating. When a location updating failure occurs, the counter

will add one. The attempt counter will be reset in the following cases:

MS powered on.





2-30

SIM inserted.

Location update successfully completed, and the service statuses switch from

Attempting to Update.

MS roaming into a new location updating area. T3212 timeout.

Location update initiated by CM sub-layer.

Attempt counter is used to decide whether to implement another attempt after T3212

timeout.

2) Location update accepted by the network

If the network accept location updating, it will send "Location Updating Accept" to MS.

When authenticating the validity of security, TMSI reallocation is a part of location

updating process. "Location Updating Accept" contains the TMSI allocated for MS and

the current LAI. In this case the network will initialize T3250.

If the network needs to prolong the RR connection so that MS can initialize MM

connection (e.g. MS sends a request subsequent to "Location Updating Request"), the

network will attach "Continue" to "Location Updating Accept" and initiate T3255. After

receiving "Location Updating Accept", MS stores LAI, terminates T3210, restarts the

attempt counter, and sets the status in SIM as Updated. If what contained in the

message is IMSI, MS will delete the corresponding TMSI stored in SIM. If the message

contains TMSI, MS will store it in the SIM and send "TMSI Reallocation Complete" to

the network. If neither of them can be received, MS will delete the original TMSI stored

in SIM. If the LAI or PLMN identifier in "Location Updating Accept" is one of "Barred

series", all of original input will be deleted. After that, MS will use "Continue" to direct it

action. If this unit exists, and MS has the underway CM service request, it will send "CM

Service Request" to the network.

3) Location update denied by the network

If location updating is denied, the network will send "Location Updating Denied" to MS.

After receiving this message, MS will terminate T3210, store the reject cause, activate

T3240, enter location denied status and wait for the network to trigger RR connection

release.

a) If the reject cause is IMSI unknown to HLR, invalid MS, invalid ME.

MS will set the location updating status as Roaming not Allowed, and store it in SIM.

Delete TMSI, stored LAI and ciphering SN and regard the SIM as an invalid one until

MS powered off or SIM removed.

b) If the reject cause is : PLMN not allow, location area not allow, international roaming

not allowed in this location area.

MS will delete any LAI, TMSI and ciphering serial key, reset the attempt counter, and

set the update status as "Roaming not Allowed". If MS receives "domestic roaming not





2-31

allowed in this location area", it will return to MM Idle and then implement PLMN

selection instead of cell selection.

Other situations will be treated as abnormal ones.

4) RR connection release after location updating

After location updating, MS will set T3240, enter "wait for network command phase"

and wait for the release of RR connection.

If MS cannot receive RR connection release command from the network within a period

of time (controlled by T3240), it will terminate RR connection. No matter RR connection

is released by MS or the network, MS will enter "idle status".

5) Abnormality at MS side

a) Access denial controlled by access level, unable to initiate location updating. MS

camps on the current serving cell, and implements normal cell reselection. Try to initiate

before denial status ends or cell changed.

b) Random access delayed (after receiving Immediate Allocation Denied): unable to

initiate location updating. MS stays in the selected cell and initializes normal cell

selection. When changing, initialize location updating before T3122 timeout.

c) Random access failed: activate T3213. Activate location updating after it times out.

d) RR connection failure: terminate location-updating process.

e) T3210 timeout: terminate location updating process and RR connection.

f) RR released before normal termination: terminate location-updating process.

g) Location update denied caused by other reasons: MS waits for RR connection

release.

For (d) ~ (g) and random access occurring for many times, MS will terminate T3210.

When T3210 times out, RR connection will be canceled, and attempt counter adds 1.

The action afterwards is decided by LAI and the record of the attempt counter:

a) The update status is "Updated", the stored LAI equals to the one received from the

previous cell, and the record of attempt counter is four. MS will maintain the "Updated"

status. The MM idle status after RR connection release is "Normal Service". MS stores

the type of location updating. After RR connection release, T3211 will be activated.

After T3211 timeout, MS will reinitiate location-updating process (adopting the stored

type).

b) If the update status is not "Updated", or the stored LAI is different from the one

received from BCCH, or the record in the attempt counter is larger than 4.

After RR connection release, MS will delete LAI, TMSI, ciphering SN in SIM, set the

update status as "Not updated", and enter MM idle sub-status "Attempt update". If the





2-32

record in attempt counter is smaller than four, T3211 stored in MS will be initiated during

RR connection release, otherwise the stored T3212 will be initiated.

6) Abnormality at NSS side

a) RR connection failure

If the RR connection failure occurs successively when there is a common program, the

network should implement according to the common program description. If RR

connection failure occurs successively and there is no common program, the location

updating process should be terminated.

b) Protocol error

If protocol error exits in "Location Updating Request", the network should return

"Location Updating Denied". The reject cause is

Mandatory information unit incorrect

Information unit not exist or unable to be realized

Invalid information unit content

Protocol error, not regulated

When these errors occurs, the network will initialize the process of channel release.

2.1.8 Access

I. Circuit service access

An MS can be either in "active" state or in "idle" state. In idle mode, MS is not allowed to

implement any transmission. In the "dedicated/active" mode, the MS can make

effective transmission to the network through an allocated channel.

In idle mode, MS gives the access cause and analysis of the cause in the 8-bit

information during access request, and gets the channel for access after channel

allocation. If the network cannot select the suitable channel type with limited cause

analysis, it will allocate an SDCCH by default. If there is no available channel during

channel allocation, the network will notify the MS to implement access attempt after a

period of time with the command "Immediate Assign Denied". After the channel

activation via Abis interface, the network sends "Immediate Assign" to MS. After

receiving "Immediate Assign", MS sets up a dedicated channel to the network with

"Setup Indication" and enters active mode. After receiving the setup indication reported

by the MS, BSC analyzes the contents of the setup indication, including the processing

of MS class mark, power control record, and encryption information. Then BSC

transmits the setup indication reported by the MS to MSC.





2-33

II. Packet service access

When no PCCCH is configured in the serving cell, packet service is accessed on the

CCCH. The BSC transfers the packet paging messages coming from the PCU to the

MS on the PCH of the CCCH. The channel request message issued from the MS is

transferred to the BTS via the RACH of the CCCH and then reported via the BTS to the

BSC. If the channel request message is for packet access (corresponding to the MOC),

the BSC will not process it and will transfer it to the PCU. At the same time, the BSC

receives the packet immediate allocation message from the PCU, transfers it to the MS

and completes the packet call access. If mobile channel request message is for Paging

Response (corresponding to the MTC), the BSC will first allocate the DCCH and enter

the active mode. On the reception of EST_IND of RR_INITIALITION_REQ message,

the BSC will transfer it to the PCU. And it will receive the PDCH message of the PCU,

and transfer it to the MS, completing packet call access.

If the MS accesses packet service via the PCCCH, then the packet call process is

transparent to the BSC. After receiving packet paging message, MS will initialize the

process of uplink Temporary Block Flow setup, and then sends the paging response

packet in data form to PCU via the air interface. PCU forwards the packet to SGSN.

After receiving the paging response, SGSN is ready to transmit downlink data.

2.1.9 Paging

Paging means that when a call is routed to the destination office, GSM/GPRS network

initializes the call at the current location area or routing area of the called MS. Packet

paging is mainly implemented at routing area, but location area is available. This is

decided by SGSN. There are two types of paging, i.e. packet paging and circuit paging,

which will be examined respectively below.

I. Packet service paging

When there are downlink data that shall be sent to the MS, SGSN needs to initiate a

packet paging call. The paging request message originated by the SGSN is sent

through Gb interface to PCU, which converts it into the packet paging request of the air

interface (Um interface) before sending. If the PCCCH channel is configured for the

BSS system, the message will be sent on the PPCH directly. If PCCCH is not

configured for the system, PCU will send the message via the Pb interface to the BSC,

which sends it on the PCH.

On receipt of the packet paging message, MS starts access procedure.

II. Circuit service paging

1) Overview





2-34

When a call reaches the MSC where the subscriber is located, the MSC sends a paging

message to all cells in that location area according to the registered location area of

MS.

In the GSM network, the concept of Location Area (LA) is introduced to reduce waste of

resources. A LA contains a small group of cells. An MS belongs to a LA at specific time.

The LA information is stored in VLR from which MSC can query them.

A paging process is completed jointly by MSC, BSC and BTS as follows:

When a call is routed to the serving MSC of the called MS, MSC first figures out the

location area of MS, and then sends the paging message to all BSCs in this location

area. The paging message contains the information that can be used to identify the

subscriber (IMSI or TMSI). BSC determines which BTS to page according to the LA,

and determines the paging channel of the MS according to the IMSI, and sends them tothe BTS. BTS will transmit the paging message of the MS on the specified PCH.

The configuration of the PCH can be changed as the traffic increases or decreases.

The PCH configuration information of each cell must be notified to each MS in the cell.

When the configuration changes, BSC must modify the broadcast messages

accordingly so that the MS in the cell can wait on the specified PCH sub-channel to

answer the paging message.

To enhance the signaling efficiency, a group of paging request combinations, called

paging group, can be sent together. A page is generally sent three times.

When flow control is allowed, the BSC can automatically adjust the configuration of

PAGCH.

If the GPRS/GSM system runs in network operation mode 1 and there exists a Gs

interface, the circuit paging of the GSM service can be sent on the GPRS channel. In

other words, if an MS is GPRS-attached, its circuit paging shall go from MSC to SGSN

and then to PCU through the Gs and Gb interfaces, and PCU will determine on which

channel to transmit the paging.

If the system is configured with PCCCH, the paging message of the circuit will be sent

directly by PCU on the PPCH or PACCH channel. If the MS is already allocated to

PDCH, it shall be sent by priority on the PCCH. If the MS is not allocated to the PDCH,

it shall be sent on the PPCH.

If the system is not configured with PCCCH, PCU transfers the paging message to BSC

through Pb interface, which then transmit the paging message on the PCH.

After receiving the circuit-paging message, MS accesses the RACH and starts the

circuit connection setup process. MS will initiate the GPRS SUSPEND process to

suspend the GPRS services and will not recover the GPRS service till the circuit is

released.





2-35

2) Paging mode

The GSM network defines three commonly used paging modes:

“Ordinary” paging mode: The paging messages are only transmitted on the

channel defined by PCH configuration and IMSI.

“Complete” paging mode: When a notice is given to an MS group in this mode, it

indicates that the paging messages of this subscriber group might be transmitted

on any PCH at the same time slot. When the PCH configuration is modified

dynamically, this mode can be used to avoid the loss of paging messages.

“Spaced” paging mode: BSS attaches a group of paging messages to another

paging channel for transmission. This is to avoid temporary overload. In other

words, the MS that receives an “ordinary” paging on the paging channel N can

receive the paging message on the paging channel N+2.

M900/M1800 BSS supports all the three paging modes: "ordinary" paging mode,

"complete" paging mode and “spaced” paging mode. Therefore, in PAGCH channel

adjustment due to traffic flow, subscribers in the serving cell will not lose the paging

message. Once a paging message is received by MS, the access allocation and

allocation initialization process is started.

If an MS is GPRS-attached in network operation mode 1, circuit paging to this MS will

go through Gs interface, Gb interface, and Pb interface, and reach the BSC by way of

MSC-SGSN-PCU. Then, there are three possibilities that the paging message will be

transmitted to the MS, which are described according to their priorities.

If the MS has been allocated with a PDCH, the message is transmitted on the

PACCH.

If the serving cell has been allocated with a PCCCH, the message is transmitted

on the PPCH.

If the serving cell is not configured with the PCCCH, the message is transmitted on

the PCH.

2.1.10 Immediate assignment

I. Overview

Immediate assignment is for the purpose of establishing the wireless connection, i. e.

RR connection with MS at Um interface. When MS needs to set up a connection, the

immediate assignment process will allocate a channel necessary for the signaling

interaction of establishing this connection. This channel can be either a SDCCH or a

TCH. Huawei BSC supports the immediate assignment of SDCCH and immediate

assignment of TCH at the level of cell.

II. Technology description

1) Channel request





2-36

The process of initialization is actually the process of random access. Whenever MS

needs to set up connection with the network, it has to send a message to the network

via RACH to request for a signaling channel. The network will decide the type of the

channel to be allocated according to the channel request. This message sent via RACHis called Channel Request. In this message, there is only 8 bits of meaningful signaling

message. In which, 3 bits are used to the minimum indication of the access cause (in

Phase 1, the cause occupies only 3 bits; in Phase 2, due to concept of half rate, the bit

occupied by the cause is not a fixed, and the maximum one can be 6 bit). Such as

emergency call, location updating, response to paging or caller request, etc. In the case

of network congestion, system can implement different processing (which type of call

will be accepted or denied) to the channel request of different access purposes

according to this rough indication, and allocate the most suitable channels for them. In

this indication, due to the capacity limit of the channel, it is impossible to transmit all

information to be transmitted, such as the specific cause of channel request, subscriber

identity and the feature of mobile equipment (all transmitted in SABM) to the network.

The other 5 bits is the identification code selected by MS at random (for Phase 1

standard). It is not used to notify the network about the MS's location but to enable the

network to identify the request initialized by different MSs. After that, the network will

send "Immediate Assign Command" (includes the information of the allocated channel)

to MS. The identification code will be returned to MS in this message. MS judges

whether the information is for it by comparing the identification code it sent and the one

returned from the network. But it has only 5 bits, which can be used to differentiate 32

MSs simultaneously. Two MSs initializing calls simultaneously do not necessarily havethe random identification codes different from each other. To further differentiate MSs

initializing calls simultaneously, the response messages on Um interface are used as

another reference. The channel request message is processed only within BSS.

All MSs with SIMs belong to a level among Level 0~9. The access level is stored in SIM.

MS can also belong to one of the 5 special access levels (Level 11~15). Such level is

also stored in SIM.

In BCCH system information, the information, such as the access levels and special

access levels allowed by the network, and whether all MS or only those of special

levels is allowed to initiate emergency calls, will be broadcast.

If the setup cause requested by MM is not emergency call, then only when MS belongs

to the access level or special access level, can its access be grated. If the setup cause

requested by MM is emergency call, then only when all MSs in the cell are allowed to

initialize emergency call, or belongs to the allowed special access level, can their

access be grated.

Since the network cannot control the access time of MS, the event of two MSs

contending for the same RACH timeslot will inevitably happen in the areas with heavy

traffic. This is called the collision. The collision leads to two results: the network will





2-37

receive a burst level from this timeslot obviously higher than the other. In this case, the

network will process the random access request with higher level. The other one is that

the network can receive neither of them due to their mutual interference. With the

increase of traffic, the possibility of loss of message due to collision will become higher.This will become the major problem of network capacity. Therefore, it is necessary to

introduce the mechanism of retranslating channel request.

MS figures out that it allowed to transmit "Channel Request" via RACH for at most M +

1 times with the following methods:

The timeslot No. Between the assign process and a "Channel Request" (not

including the timeslot containing the information itself) is selected at random from

{0, 1 MAC (T, 8)-} with the same probability.

The timeslot No. between the two consecutive "Information Request" of MS is

selected at random from {S, S + 1, , S + T – 1} with the same probability.

T is the parameter "Tx integer" broadcast on BCCH; M is "Max Retrans"; the value of S

depends of the configuration of CCCH. See Table 2-4.

Table 2-4 Value of Parameter S

Tx Non-combined CCCH Combined CCCH/SDCCH

3, 8, 14, 50 55 41

4, 9, 6 76 52

5, 10, 20 109 58

6, 11, 25 163 86

7, 12, 32 217 115

If the immediate assign command is not received even after Max Retrans, MS will

return to idle mode.

After transmitting initial channel request, MS will activate T3120 and stay on the entire

downlink CCCH (to receive answer) and BCCH.

When T3120 times out while RACH retransmission times has not exceeded "Max

Retrans", MS will retransmit the channel request message containing a new random

reference, and activate T3120 with a new value.

When T3120 times out, and the Max Retrans is reached, MS will activate T3126, and

then wait for a period of time and allow network to give up. If no network response

received after T3126 timeout, MS will give up request attempt and perform cell

reselection.

2) The allocation of the initial channel





2-38

After correctly decoding the Channel Request of MS, BTS will send Channel Required

to BSC via Abis interface. This message contains important attachment information

and the estimation to TA that is important to activating timer advance control. After

receiving this message, BSC will select a corresponding idle channel for MS accordingto the judgement to the existing radio resources. However, the availability of the

allocated channel and the related terrestrial resources is to be acknowledged with the

response from BT. This process is realized by sending "Channel Active" from BSC to

BTS to query the availability of corresponding terrestrial resources (e. g. transmission

circuit). This message indicates all properties needed in activating the channel,

including channel type, working mode, physical feature and initial lead. When the

corresponding resources are ready, BTS will return "Channel Active ACK" as a

response to BSC.

a) Immediate assignment

After BSC receive Channel Active ACK from BTS, will send Immediate Assignment or

Extended Immediate Assignment to allocate dedicated signaling channel for MS in the

non-acknowledge mode, via the CCCH for MS receiving Channel Request. Immediate

Assignment contains the assignment information of only one MS, while Extended

Immediate Assignment contains the assignment information of two MSs. BTS can send

Immediate Assignment or Extended Immediate Assignment on any message block of

downlink CCCH. Therefore it is necessary for MS to monitor all information block on

CCCH. The allocated channel type (TCH or SDCCH, channel mode is set as signaling)

is decided by the carrier.

Normally, if there is an idle SDCCH available that can satisfy the access request, BSC

will allocate SDCCH. The process of requesting for SDCCH connection includes

location updating, IMSI detach, supplementary service, short message at non-session

status and services only supported by SDCCH. MS initializes access request, and BSC

allocates a SDCCH for this call. This channel seizes 1/8 sub-timeslots of a timeslot. The

signaling interaction necessary for call establishment is implemented on that channel.

The signaling flow of SDCCH immediate assignment is illustrated in Figure 2-7.





2-39

SDCCH: Assignment Command

SDCCH: Ciphering Mode Complete

SDCCH: Authentication Response

SDCCH: Authentication Request

MS BTS BSC MSC

Channel Request

Ciphering Mode Command

Channel Required

Channel Active

Channel Active ACK

Immediate AssignmentImmediate Assignment

SDCCH: SABM

SDCCH: UAEstablishment Indication

Complete Layer3 Informaiton

Encryption CommandSDCCH: Ciphering Mode Command

SDCCH: Setup

SDCCH: Call Proceed Assignment Request

Channel ActiveChannel ACKAssignment Command

TCH: SABMTCH: UA

TCH: Assignment Complete Assignment Complete

TCH: AlertTCH: ConnectTCH: Connect ACK

Figure 2-7 Immediate assignment

If TCH has been allocated before immediate assignment, there is no need to reallocate

TCH during the process of assignment. Mode conversion process can be used to

change the function of TCH from signaling transfer to voice transmission. Signaling

flow of TCH immediate assignment is illustrated in Figure 2-8.





2-40

TCH: Ciphering Mode Complete

TCH: Authentication Response

TCH: Authentication Request

MS BTS BSC MSC

Channel Request

Ciphering Mode Command

Channel Required

Channel Active

Channel Active ACK

Immediate AssignmentImmediate Assignment

TCH: SABM

TCH: UAEstablishment Indication

Complete Layer3 Informaiton

Encryption CommandTCH: Ciphering Mode Command

TCH: Setup

TCH: Call Proceed

Assignment RequestMode Modify

Mode Modify ACK

Assignment CompleteTCH: AlertTCH: ConnectTCH: Connect ACK

Channel Mode ModifyTCH:

TCH: Channel Mode Modify ACK

Figure 2-8 Immediate assignment

Messages of immediate assignment or extended immediate assignment contain:

Description of assigned channel.

Information field of channel request and abbreviated frame No. of the received

channel request frame (abbreviated frame No. is a frame No. with narrow value

range calculated from the TDMA frame No. received by BTS during channel

request.)

Initial lead; Start time indication (optional).

The random reference and abbreviated frame No. are directly related to the MS

channel request. They are used to reduce the conflict of request among MSs. TA is the

initial lead calculated from equalizing the channel request information received by BTS

on RACH. MS figures out the next initial lead for transmitting according to TA.

After receiving immediate assignment or extended immediate assignment, MS

switches to the channel assigned by the network, sets the channel modes as signaling

only and sends the SABM with information field via the allocated channel to establish

the main signaling link.





2-41

Immediate assignment or extended immediate assignment message, containing the

start time and description of possible alternative channel, can be used to indicate the

frequency change in the process.

If the received immediate assignment or extended immediate assignment message

contains only the description of the channel used after start time, MS will access to the

channel during the time waiting for start. If it misses the time, MS will immediately

switch to this channel after receiving the message.

If the message contains the description of the channel used after indication time as well

as that before the indication time, MS will access the channel after receiving the

message. If MS is ready for access to the channel upon the indicated time, MS will first

access to the channel used before the indication time, and switch to the one after the

indication time when the time comes (new frequency series, MAIO and HSN). If MS is

ready after the specified time it will access the channel after the indication time.

If MS has already sent the channel requests for maximum allowed times RR entity will

start T3126. After T3126 timeout, immediate assignment process will be terminated. If

the random access program is initiated by MM, system indicates failure of random

access to MM.

After sending the first channel request message, MS starts to monitor the system

message on BCCH as well as the CCCH timeslot corresponding to its paging group, i. e.

immediate assignment command may appear in any CCCH message block in 51

multiframe. Therefore it is necessary for MS to monitor the entire CCCH block after sending channel request, i. e. decode the messages of the entire paging sub-channel

for response from the network.

If the network adopts FH, MS will decode MA with the CA got from BCCH system

message. CA refers to all the frequencies used in the cell (including FH frequencies).

MA refers to all FH frequencies used in the cell.

b) Immediate assignment denied

If there is no available channel for BSC to allocate, the network can send the immediate

assignment denied message in non-acknowledge mode to MS via CCCH. The rejectcause can be MSC traffic closed, radio resources shortage, TA value exceeding limit,

channel activation no response and BSC traffic overload. But system does not specify

the part on downlink CCCH for immediate assignment denied transmission. The

message of immediate assignment denied contains request reference and waiting

indication.

After receiving Immediate Assignment Denied, as the response to one of the last three

channel requests, MS terminates T3120, activates T3122 with the specified value and

returns CCCH idle mode. MS cannot start RR connection attempt until T3122 timeout.

MS is not allowed to initialize another call attempt except for emergency calls until





2-42

T3122 timeout. Emergency call attempt can be established in the same cell before

T3122 timeout as long as no "Immediate Assignment Denied" of RR emergency

attempt received.

It corresponds to the immediate assignment extension. In order to improve AGCH

efficiency, the format of extended immediate assignment denied is introduced. The

message of extended immediate assignment denied can contain information of

rejecting at most four MSs.

The value of the wait indication information unit (T3122) depends on the cell receiving

this message.

After T3122 timeout, MS will not respond to paging but sends "Channel Request"

instead till MS receives "Paging Request".

c) Signaling channel assignment overlap

The system may have a slow response to the channel request of MS, which results in

request retransmission. In this case, system do not know whether a channel request

message is a retransmitted one, so it may send the immediate assignment command to

the MS for multiple times. MS will use the channel in the first assignment message it

decoded. The others are regarded invalid ones. But according to the specification, MS

should receive the last three network response messages to the channel request. This

is called allocation overlapping. It is possible to cope with CCCH congestion caused by

to many overlapped allocations by reducing the retransmission of MS or shorten T3101.

This measure can avoid the waste of system resources.

3) Initialization message

After receiving immediate assignment command, MS will decode this message. If the

random identification code and the abbreviated frame No. satisfy the requirement, MS

will tune its transceiver equipment to the specified channel and start to transmit

signaling according to TA specified by BSC and maximum transmitting power (defined

in the parameter "MS TxPWR MAX CCH" in BCCH system broadcast message). The

first task for MS on the allocated SDCCH/TCH is to send a SABM frame to establish

asynchronous balance mode (service access point type: SAPI = 0) so as to establish

signaling message link connection in acknowledge mode. In GSM specification, SABM

has a signaling message, i. e. initialization message. On Um interface, SABM frame is

a message requesting for the establishment of a multiframe response operation mode

on LAPDm. This message contains the L3 service request message. The reason for

different standards about standard HDLC is to guarantee the correctness of MS

receiving. If two MSs send the channel requests with the same message content at the

same (possible in the case of high load), BSS will repines to one of them only. While

these two MSs can both be allocated with the same dedicated channel. To settle this

problem, there should be a mechanism judging such contention. According to the

specification, the cell will send a UA frame (no No. verification) with the content





2-43

completely the same as that of SABM frame after the cell has received the SABM frame.

MS compares it and the SABM information. If the content is completely the same, the

access will proceed. Otherwise, it will give up this channel and repeat the immediate

allocation process. Only when the consistency is guaranteed, will MS stay on thechannel.

According to different request causes, the initialization messages in SABM can be

divided into four types: CM service request (call establishment, short message,

supplementary service management), location updating request (generic location

updating, periodic location updating, IMSI attach), IMSI detach and paging response.

All these messages contain the identity of MS, detailed access cause and Classmark of

MS (used to indicate some key features of MS, such as transmission power level,

ciphering algorithm, short message capability and frequency capability).

Upon receiving SABM frame, BTS will send a message "Establishment Indication" to

BSC. On Abis interface, this message is used to notify LAPDm that the connection has

been established. It is a response to the immediate assignment message. After

receiving the indication message of establishment, BSC will send a L3 service request

message (Complete Layer3 INFO) to MSC. To be specific, this message is Location

Updating Request, CM Service Request, Paging Response and IMSI Detach. This

message contains the SCCP connection request (SCCP CR), cause of CM service

request (e.g. MO call, emergency call, location updating and short message service),

ciphering key sequence No., LAC, CI, physical information of this MS (e.g. transmitting

power level, ciphering algorithm support, pseudo-synchronous capability and shortmessage capability) and the ID of MS.

Although the MTP connection at An interface has been established before the session,

there should still be a SCCP connection on L2 for each call. This establishment request

message will be transmitted in the SCCP CR message via A interface. If the request is

permitted, the first downlink message at An interface will be contained in the CC frame

at SCCP layer. For SCCP layer, the exchange between CR and CC is the exchange

between original reference address and destination reference addresses. For different

calls, the same SPC may refer to different original addresses and destination

addresses. If SCCP cannot be established, MSC will send the message SCCPRefused. The access ends at this step. The signaling link between MS and MSC has

been established, MSC at this phase is able to control the transmission feature of the

RR management, and BSS is in the status of monitoring transmission quality and ready

for handover.

4) Phase1 and Phase2 MS

BSC cannot differentiate whether a call is for voice, data or signaling completely

according to MS establishment cause. In the case of Phase 2 MS, BSC can obtain the

access request cause more detailed than that of Phase1 MS. For Phase 2 MS, BSC is

able to recognize the information unit "Channel Needed" in the paging message. This





2-44

information unit indicates whether the current channel is for voice/data or signaling. MS

selects a suitable establishment cause to response according to its own capability.

Huawei BSC supports the information unit "Channel Needed" in the paging message.

III. Parameter

Huawei BSC controls the function of immediate TCH assignment with the switch of

"Immediate assign TCH. The detailed configuration process, as well as the data table

and parameters involved are listed below:

[Cell/Modify Cell's Call Control Parameter/Modify Cell Call Parameter/Call

Control]

Parameter: Immediate Assignment of TCH

If “Immediate Assignment of TCH” is set as “No”, this indicates that the function of

immediately assigned TCH is disabled. All call access requests use SDCCH.

If “Immediate Assignment of TCH” is set as “Yes”, this indicates that the function of

immediately assigned TCH is enabled. For emergency call and call re-establishment,

BSC will preferentially assign TCH for them. If no idle TCH is available, the BSC

assigns SDCCH for them. For other the channel access request of calls, SDCCH will

be preferentially assigned and then TCH.

2.1.11 Assignment

I. Overview

BSS switches MS to TCH by means of assignment. Normally, the assignment is

finished at the cell where the call is initialized. Huawei BSC supports the function of

direct retry, which can assign MS to other cells.

Huawei BSC supports re-assignment process. When BSC receive ASSIGNMENT

FAILURE from UM interface, BSC allocate a new radio channel and initiate the second

assignment process.

II. Working principle

After MS initializes service request, BSC will assign the MS to TCH by means of the

assignment process. If BSC figures out that there is idle TCH in the cell where MS

initialized the call, it will assign the MS to that TCH. Huawei BSC provides two

algorithms of channel selection: "Huawei Channel Algorithm I" and "Huawei Channel

Algorithm II". The channel allocation algorithms of Huawei guarantee that the currently

allocated channel is the best one.





2-45

If BSC has already assigned the MS to TCH during immediate assignment, it will not

assign the MS to a new TCH, but to the old one.

If there is no idle TCH in the cell of the MS, the function of directed retry can be used to

assign the MS to other cells with idle TCH and go on with the service. BSC can select

the best cell among the adjacent cells with the measurement reports as the destination

cell in directed retry.

If BSC receive ASSIGNMENT FAILURE from UM interface, the re-assignment process

is controlled by the parameter "Allow Reassign". Re-assignment process chooses

different TRX preference. If there is no other idle TRX in the same cell, the

re-assignment process is initiated at the same TRX.

III. Parameter

BSC decides whether to use the function of directed retry with the parameter "Directed

retry permitted".

The detailed data table and parameters involved are listed below:

[Cell/Modify Cell's Call Control Parameter/Modify Cell Call Control

Parameter/Call Control]

Parameter: Directed Retry Perm

Parameter: Allow Reassign

2.1.12 Authentication

One of GSM system's advantages comparing with analog system is security system. It

has the following improvements: on access network: AUC authenticates the subscriber;

on radio path: communication information ciphering; EIR identifies the mobile

equipment; IMSI is protected by TMSI; SIM is protected with PIN.

The authentication process is one of the common processes of Mobility Management

(MM) process. The common processes of MM includes authentication process,

identification process, TMSI reallocation process and IMSI detach process initialized by

MS. Other common processes will also be mentioned in this chapter.

I. Authentication process

1) Authentication triplet

Authentication and ciphering process is realized with the triplet allocated by the system.

The triplet is generated in the Authentication Center (AUC). After subscribing to GSM

service, each MS will be allocated with a MSISDN and IMSI. IMSI is written to the SIM

of the subscriber with SIM writer. Together with this IMSI, the authentication key Ki

uniquely corresponding to the IMSI is also stored in the SIM and in AUC. In AUC, there





2-46

is a pseudo-random code generator used to generate a unpredictable pseudo-random

code RAND (randomly selected from 0~2128

-1). The GSM specification also defines the

algorithms of A3, A8 and A5 used in authentication and ciphering process. In AUC,

SRES is generated by processing RAND and Ki with A3, and Kc is generated byprocessing RAND and Ki with A8. RAND, Kc and SRES make up the triplet, which will

be transmitted to HLR and be saved in the database of that subscriber. Normally, AUC

transmits five triplets at a time to HLR. HLR can store 10 triplets. When MSC/VLR

requests HLR for triplet, HLR will transmit five triplets to it. MSC/VLR uses one triplet

each time. When there is two triplets left, it will request HLR for triplets again.

Below is the detailed introduction to the process of parameter transference.

2) Authentication process

There are two purposes for authentication: one is to check whether the identification

provided by MS is effective, and the other is to allocate a new ciphering key for MS.

After the establishment of RR layer between MSC and BSS, the network is able to

decide whether to trigger the authentication process to verify the identification of the

mobile subscriber. Whether to trigger the authentication process depends on Kc at

network side (stored after the previous processing of the MS service) is the same as

that stored in the MS accessing currently. If they are the same, system will skip

authentication and go to ciphering process with Kc stored in MS. Otherwise, Kc has to

be calculated with authentication process.

To enable the ciphering in the case of initializing RR connection without authenticationprocess, the concept of ciphering key sequence number is introduced. It is called

CKSN in the specification. CKSN is stored in SIM as well as in MSC/VLR together with

Kc and is processed by the network. In the first L3 message (e. g. location updating,

CM service request, paging response, MS will indicate the CKSN to the network. CKSN

= 0 means no Kc allocated. The calculation of Kc is illustrated in Figure 2-9.

Ki RAND

A8

Kc

Figure 2-9 Kc calculation

After receiving Complete Layer3 INFO, MSC will send "Process Access Request" to

VLR for authentication and ciphering. VLR will return the message of "Process Access





2-47

Accepted". And then, MSC/VLR will send the message “Authentication Request” to MS

to trigger the authentication process. At the same time, T3260 will be activated.

"Authentication Request" contains a RAND and a CKSN. RAND is 128 bit. MS will send

the message "Run GSM Algorithm" to SIM after receiving this message. A 32-bit SRESwill be generated by processing Ki stored in SIM and this SRES with A3. Meanwhile, a

64-bit Kc is calculated by processing Ki and RAND with A8. MS will store it and CKSN

to the suitable position SIM for the future activation of ciphering transmission.

If RR connection exits, MS should respond to the authentication request message. MS

will sends the SRES to the network with the message Authentication Response. After

receiving "Authentication Response", the network will terminate T3260 and check the

validity of the SRES. Since Ki is stored in VLR or HLR as a subscriber data, the A3 and

A8 will also be carried out to generate a SRES and Kc and store them in VLR. System

compares these two SRESs. If they are the same, the authentication will succeed andaccess to network will be grated, see Figure 2-10. And after that MSC proceeds ahead

with the ciphering process. If they are different, authentication failed, and system will

reject the access of the MS. The authentication process ends at this step.

Ki RAND

A3

SRES

KiRAND

A3

SRESEqual

Authenticationsucceeded

AUC

MS Network

Figure 2-10 Authentication algorithm

A3 and A8 can be executed either in MSC/VLR or in HLR/AUC. But it will be

complicated for MSC/VLR, but simple for HLR/AUC for it stores Ki. Furthermore, it is a

better way to achieve security and roaming. However, it causes the increase of

signaling traffic between HLR and MSC, for each authentication, HLR/AUC will send

RAND, SRES and Kc to MSC/VLR.





2-48

3) Unsuccessful authentication

If the authentication failed, the network can use the subscriber's identification.

a) MS uses TMSI

If MS uses TMSI, the network can initiate the identification process. If the TMSI does

not correspond to the IMSI, authentication will be restarted.

b) MS uses IMSI

If MS uses IMSI or the network decides not to initiate identification program, then

"Authentication Denied" will be sent to MS. After sending this message, all MM

connection in process will be released, and then the network will initialize the RR

connection release process.

After receiving "Authentication Denied", MS will set the status of SIM as "Roaming

Denied", and delete the existing TMSI, LAI and CKSN, and regards SIM invalid until MS

powered off or SIM removed.

4) Abnormality handling during authentication process

a) RR connection failure

If RR connection is detected before receiving "Authentication Response", the network

will release all MM connection and terminate all running MM special process.

b) T3260 timeout

If T3260 timeout, the network will release RR connection. In this case, the network will

terminate the authentication process and all ongoing MM process, releases all MM

connection and initializes RR connection process.

2.1.13 Ciphering

I. Overview

The feature of wireless transmission has a negative effect on the security and interest

of the subscribers. The analog mobile communication has always been the victim of

interception and misappropriation. The digital transmission of GSM guarantees

excellent security. The encryption function deals with the security for information

exchange between MS and BTS, including signaling information and user information.

It is up to the radio resources management to decide whether to adopt the encryption

mode or not. The encryption function is implemented in the BTS to encrypt user data.

The related parameters must be sent to the encryption program. The ciphering key Kc,

generated by AC and stored in the MSC/VLR, is sent to the BTS before encryption

starts.





2-49

II. Technical description

In order to achieve a general understanding of the encryption/decryption process of

GSM, we will examine it here from three perspectives: TMSI, encryption process and

Kc generation.

1) TMSI (Temporary Mobile Subscriber Identity)

IMSI is the identity for mobile subscriber. Due to the importance of IMSI it is not

transmitted on the radio link repeatedly. VLR allocates a TMSI to the subscriber during

the MS registration. Afterwards TMSI is used in place of IMSI to protect the IMSI for the

sake of subscriber security.

The relation between TMSI and IMSI is not fixed. TMSI is valid only in a VLR area.

2) Ciphering and deciphering processes

a) Initializing ciphering mode setting

After authentication process, MSC will send "Ciphering Mode Command" to BSC. This

message contains Kc. BSC sends Ciphering Mode Command to MS to indicate

whether ciphering is necessary, and if needed, which type of dedicated resources are

to be adopted.

b) Ciphering mode setting complete

Once MS receives the valid "Ciphering Mode Command", it will send the Kc stored in

SIM to the mobile equipment. The valid "Ciphering Mode Command" are as follows:

"Initialize Ciphering" is indicated when MS is in the status of "Non-ciphering".

"Non-ciphering" is indicated when MS is in the status of "Non-ciphering".

"Non-ciphering" is indicated when MS is in the status of "Ciphering".

MS will regard the "Ciphering Mode Command" of other type received as an incorrect

one. It will respond with "RR Status", and the cause value is "Error: protocol not

defined".

After receiving the indication of "Ciphering Mode Command" and the ciphering process,

MS should initiate the Tx and Rx in ciphering mode. After MS has activated the actions

of "Ciphering Mode Command", it will returns "RR Ciphering Mode Complete" to the

network. If the field "Ciphering Response" in the information unit of ciphering response

message "IMEI shall be included", then MS will include its IMEI in "RR Ciphering Mode

Complete".

After receiving "Ciphering Mode Complete", the network will initialize the transmission

in ciphering mode.

BTS and MS carry out the encryption/decryption of the radio path. The process of

encryption and decryption is shown in Figure 2-11.





2-50

Kc

Mod 2+1

A5

TDMA（

Data flow

Data not ciphered

Mod 2+1

A5

TDMA（ Kc

Data not ciphered

Data flow Data flow

TX RX

Figure 2-11 Process of ciphering and deciphering

The algorithm for generating the ciphering code is called A5. Using the Kc consistent in

MS and the network (64 bits) and the current pulse string frame number (22 bits), it

calculates result is the 114-bit ciphering sequence (the data flow in Figure 2-11), which

then performs "exclusive or" operation together with the burst 114 bit (the data not

ciphered in Figure 2-11). The frame No. code consists of three values (T1, T3 and T2).

If the communication lasts as long as the period of hyper frame (about 3 and half hours),

the ciphering sequence will appear repetitiously.

On uplink and downlink, the network uses the same ciphering sequence. For each

burst, one sequence is used for the ciphering in MS and the deciphering sequence of

BTS, the other one is for the ciphering of BTS and deciphering of MS.

According to the system configuration, MS can decide whether to report the processing

power of MS after Authentication Request. The name of this message is Classmark

Change. Its content is the same as that in the establishment indication, and is more

detailed in the description of ciphering algorithm at MS side. The establishment

indication states whether A5/1, A5/2 and A5/3 are supported, while the Classmark

Change further states whether A5/4~A5/7 are supported. After receiving this message,

the network first response with the message MS PWR CTRL to describe the power

range available for MS and the transmitting power of the TRX corresponding to this MS.

3) Generation of Kc





2-51

A8

storing KC

RAND generator

A8

storing KC

MS NSS

RAND

Kc

Ki Ki

Kc

Figure 2-12 Generation of Kc

Ciphering key (Kc) is produced by A8 algorithm as shown in the Figure 2-9. Here Ki is a

user authentication key. After registering in the network, a subscriber obtains the Ki,

which is stored in the authentication centre and the SIM card.

The MS and the network use the same Ki and random number (which is generated by

the network and transmitted to the MS) so that the same Kc can be obtained.

III. Parameter

Condition:

MSC support ciphering, all kinds of ciphering algorithm and authentication in service

access procedure.

Huawei BSS supports both A5/1 and A5/2. The data configuration involved is as

follows:

1) BSC [BSC/Modify BSC Interface Phase Flag]

A Interface Version: GSM_Phase_2

Um Interface Version: GSM_Phase_2

Abis Interface Version: GSM_Phase_2

2) BSC [Cell/Modify Cell's system information/Modify Cell Configuration Data]

Encryption Algorithm Setting: Encryption Not Support, A5/1, A5/2





2-52

Note:

This manual only introduces the related parameters. All data configuration of BSC is realized with auto

data configuration console. For details, see M900/M1800 Base Station Controller Data Configuration

Manual.

For the configuration of ciphering algorithm, it is recommended select ciphering option not selected for

BSC. This is because BSC software selects ciphering algorithm from the intersection of algorithms

allowed in MSC data configuration, algorithms allowed in BSC data configuration and algorithms

supported by MS. If the intersection contains multiple algorithms, the one with the largest algorithm No.

will be selected. The meaning of algorithm No.: 1 (No ciphering), 2 (A5/1), 3 (A5/2)… 8 (A5/7).

2.1.14 DTX

I. Overview

In the process of communication, only 40% of time of the mobile subscriber is engaged

in session. Most of the time is not engaged in the transmission of voice message. If all

information during the non-session period is sent to the network, not only the system

resources are wasted, but also the intra-system interference will be worsened.

To tackle the above problems, GSM adopts Discontinuous Transmission (DTX). When

there is no session, the transmitting channel is closed to lower the interference leveland improve the system efficiency. In addition, this function also saves the power

consumption of MS. When transferring data, this function cannot be applied. DTX

affects the transmission of TCH frame.

There are two types of voice transmission in GSM system: one is normal mode, the

voice stream is encoded as 13kbit/s regardless of the MS's session status, and the

other is DTX mode. Only one mode can be selected in one session. When both parties

of the communication are GSM subscribers, DTX will have a negative effect on the

communication quality. Therefore, DTX mode is not allowed on the occasion.

II. Technologies description

If DTX mode is adopted, the voice coding of 13 kbit/s will be used in voice activation

period, and that of 500 bit/s (for transmitting the feature parameter of comfort noise only)

in non-voice activation period.





2-53

BTS

SP

Infor-

mation

VAD

Voicecoding

SID

DTXMSC

Information

BFI

SID

TAF

DTX

Proce-

ssing

Voice frame

replacing

Voicedecoding

CN

MSIRAU

Proce-

ssing

VAD: Voice Activity Detection SID: Silence Indicator CN: Comfort NoiseTAF: Time Adjust Flag BFI: Bad Frame Indication SP: Speech Flag

Figure 2-13 Principle of DTX

1) Voice Activity Detection

Voice Activity Detection (VAD) indicated the time to use DTX. When DTX is activated, it

is used to detect whether voice or noise is transmitted.

VAD algorithm and voice coding/decoding algorithm is closely related. This algorithm

judges whether voice or noise is contained in the output frame by comparing the filter

signal and the configured threshold. It also indicates whether the auxiliary bit of this

frame is transmitted. This judgement is based on the principle of the energy of noise

being lower than that of voice.

VAD generates a group of threshold values in each voice block of 20 ms for judgingwhether the next voice block of 20 ms is voice or noise. If the background noise is too

loud, it will be transmitted as the voice signal.

2) Silence descriptor

The noise coding process is similar to that of voice coding process: after sampling and

quantization, each 20 ms will be encoded as a noise block. The encoded noise block

will also become a block of 260 bits like the voice block. This is a Silence Descriptor

(SID). SID frame is applied to channel encoding, interleaving, ciphering and modulating

like voice frame to become a field containing noise message and be transmitted in the 8

consecutive bursts.

A complete SACCH message block on TCH has four 26-multiframe (480 ms). To

enable the peer end to differentiate the voice frame and SID frame, these 8 consecutive

bursts are fixedly arranged at the beginning of the third multiframe. Other frames

(excluding SACCH) within the same period will not be used to transmit any message.





2-54

Caution

The SID frame generated from 20 ms noise block completes the process of interleaving together withthe SID frames before and after it.

The first SID frame completes the interleaving together with the voice frame before it and the SID

frame after it.

The DTX functions are optional, independent in the direction of uplink/downlink and

based on the control unit of a cell. The uplink and downlink DTXs are two processes

independent to each other, and are activated by system parameters respectively.

There measurement methods in GSM system:

Global measurement: average the levels and quality of the 100 timeslots within the

entire period (totally 4 TCHs of 26-multiframe, idle frame not included)

Partial measurement: average the levels and quality of 12 timeslots, including 8

consecutive TCH bursts and 4 SACCH bursts containing measurement report. To

ensure the consistency, no matter whether the uplink or downlink activates DTX,

BTS and MS should both complete these two types of measurement. Since each

SACCH measurement report of BTS and MS indicates whether DTX is used, BSC

can select whether to use global measurement or partial measurement to judge

according to according to the measurement report.

Note:

No matter whether DTX is used, the uplink and downlink will proceed with global and partial measurement.

DTX, which is applicable for voice and non-transparent data transmission, involves the

operations of MS and TRAU.

No matter whether DTX is used or not, the decision-maker is MSC and the executor is

BSC.

III. Parameter

[Cell/Modify Cell's System Message/Modify Cell System Message/Bsic Data]

Parameter: Discontinuous Transmission Indication





2-55

2.1.15 Frequency hopping

I. Overview

The purpose of frequency hopping (FH) is for obtaining better security and

anti-interference capability. There are fast FH and slow FH. Fast FH means that the

change rate of frequency is faster than modulation rate of signal. In GSM system, it is

required that the frequency should remain unchanged within a burst period. Therefore,

the FH in GSM system belongs to slow FH. It involves frequency diversity and

interference diversity technologies. The frequency occupied by channel in the Um

interface of GSM system is changed regularly. The frequency of changing frequency is

about 217 times per second.

The FH can avoid the attenuation caused by multi-path transmission and samefrequency interference, and improve the average C/I of the interference restriction

system (especially in cities), thus greatly improving the quality of session,

strengthening the capability of high-density multiplexing and increasing the system

capacity. Adopting FH can improve the transmission quality of the slowly moving MS by

6, 5 dB. Besides, FH can also improve the security of communication.


1) FH modes

FH means that the carrier containing meaningful information hops under the control of asequence. This sequence is called frequency-hopping sequence (HSN). An HSN is an

array of all frequencies in a frequency set uniquely defined with HSN, Mobile Allocation

Index Offset (MAIO) and Frame No. (FN) by using certain algorithms. Channels on

different timeslots (TN) can use the same HSN. Different channels on the same

timeslot in the same cell should use different MAIOs.

FH mode can be divided into frame FH and timeslot FH by the concept of time-domain.

FH mode can be divided into RF FH and base band FH by carrier realization mode.

Huawei BSS realizes the base band FH at timeslot level, RF FH at timeslot level, baseband FH at frame level and RF FH at frame level.

2) Frame FH

Frequency changes for each TDMA frame. In the mode, each carrier can be regarded

as a channel. The TCH of the TRX which bears BCCH cannot be used for FH while

other different TRXs should have their own MAIO. This is the expiation of timeslot FH.

3) Timeslot FH

Frequency changes for each timeslot of each TDMA frame. The TCH of the TRX which

bears BCCH can be used for FH. But currently it is realized only on the occasion of

base band FH.





2-56

4) RH FH

Both Tx and Rx can be both used in FH. In a cell, the number of FH frequencies

engaged in FH can be larger than the number of TRX.

The RF FH of the M900/M1800 BTS is enabled through real-time switchover between

two frequency synthesizers. There are two advantages for this implementation: first,

lower speed requirements of the frequency synthesizer can be practical, i. e. the speed

requirements are easier to implement; second, one of the two frequency synthesizers

serves as the standby when there is no FH to enhance system reliability.

Huawei BTS adopts dynamic loop bandwidth and Ping-Pong handover to solve the

inconsistency between fast FH and signal quality, and realize the unrestricted FH in

GSM 900 bandwidth of 25 MHz and DCS 1800 bandwidth of 75 MHz. All FH indices

satisfy the requirements in GSM protocols.

Dynamic loop band width technologies: local oscillation signal is mainly decided by

reference clock (phase discrimination frequency), voltage controlled oscillator and loop

bandwidth, etc. The phase noise of local oscillation within the loop bandwidth is decided by

reference clock, and that beyond loop bandwidth is decided by collage controlled oscillator.

During the operation of Huawei BTS, loop bandwidth needs to be dynamically adjusted

along with the needs of system. If the system is not in the working status, loop

bandwidth changes back to best bandwidth, so that the output signal can be the best,

and the best performance of the system can be guaranteed.

Ping-Pong handover: Two identical oscillators are designed on the circuit. A switch is in

charge of selecting between these two oscillators. When one oscillator is working, the

other one locks on the next frequency quickly. Switching to another oscillator is realized

with a switch between two timeslots. This avoids the instant performance worsening at

the beginning and end of the timeslot.

5) Base band FH

Each transmitter works on a fixed frequency. Tx is not involved in FH. The transmitting

FH is realized by switching the base band signal. Rx is involved in FH. Therefore the

number of FH frequencies in a cell cannot be larger than number of the TRXs of the cell.

When a TRX is faulty, the system can skip it when implementing FH.

For base band FH, the parameter “TRX Aiding Function Control” in the [cell

configuration table] must be set to “Allowed & Recover When Check Res.”.

When base band FH is opened, if some TRXs in the FH group are faulty, the serious

voice disconnection, caused by frame lost, will appear during call. After “TRX Aiding

Function Control” is set to “Allowed & Recover When Check Res.”, the system will

close base band FH function in this case. When those TRXs recover, system will

activate FH on resource checking.





2-57

If the base band FH must be opened, the faulty TRXs must be removed from the FH

group.

Huawei BTS adopts the technology of (FH_BUS), which implement FH on the basis of

timeslot exchange. Each transmitter is tuned to a fixed frequency, and has a fixed ID.

The coder of transmitter encodes the downlink signal to convert the data to burst format.

It calculates the channel (i. e. TRX) to be modulated for the burst according to FH

algorithm, and adds the attached information related to power control to generate a

special data packet. The coder transmits the data packet periodically (sub-timeslot).

Modulator checks the TRX ID of the data packet from each sub-timeslot. If the TRX ID

is different from the local TRX, it will receive that from the next sub-timeslot. If the TRX

IDs are the same, it will accept the data packet, and delay for a timeslot and then

transmitted to the air interface. Base band FH has a very high requirement on the

real-time identification of the ID of TRX. Huawei base band FH technology realizes fastand reliable TRX ID identification on the basis of the ASIC.

6) FH algorithm

Parameters involved:

CA: Cell allocation table, i. e. the collection of frequency ID used in the cells.

FN: TDMA frame No., broadcast on the synchronous channel. BTS and MS

achieve synchronous with FN (0~2715647).

MA: the radio frequency ID collection for MS FH, a subset of CA. M contains N

frequency Ids, 1 ≤ N ≤ 64.

MAIO: Mobile Allocation Index Offset (0~N-1). During communication, the radio

frequency ID adopted on air interface is an element in MA. MAI (Mobile Allocation

Index, 0~N-1): indicating an element in MA, In other words, the frequency actually

used is decided by MAI. MAIO is an initial offset of MAI. Its purpose is avoid

multiple channels contends the same carrier.

HSN: FH serial No. (generator) (0~63). If HSN = 0, it will be cycle FH, and if HSN ≠

0, it will be random FH.

The process of calculating the actual working frequency on each FH timeslot is as

shown in Figure 2-14.





2-58

FNT2(0~25)

FNT3(0~50)

MAI

(m0~mN-1)

MAIO

(0~N-1)

Represent

in 7 bit

T1R=

T1 MOD 64

Exclusive OR

FNT1(0~2047)

HSN

(0~63)

Addition

Look-up table

Addition

M'=M mod 2^NBINT=T3 mod2^NBIN

M'<N

S=M'S=(M'+T) mod N

MAI=(S+MAIO) mod N

RFCN=MA(MAI)

7bit

5bit11bit

6bit

6bit

7bit

7bit

8bit

6bit6bitNBIN bit

NBIN bit

YN

NBIN bit

NBIN bit

NBIN bit

Figure 2-14 FH algorithm

In the figure above:

MAI = (S + MAIO) MOD N (S is the result after calculating the frame No.)

RFCHN = MA (MAI)

mod: MOD

^: power

NBIN: INTEGER(log2N + 1)

Table: see Table 2-5.





2-59

Table 2-5 "Table" meaning table

Address Content

000~009 48 98 63 1 36 95 78 102 94 73

010~019 0 64 25 81 76 59 124 23 104 100

020~029 101 47 118 85 18 56 96 86 54 2

030~039 80 34 127 13 6 89 57 103 12 74

040~049 55 111 75 38 109 71 112 29 11 88

050~059 87 19 3 68 110 26 33 31 8 45

060~069 82 58 40 107 32 5 106 92 62 67

070~079 77 108 122 37 60 66 121 42 51 126

080~089 117 114 4 90 43 52 53 113 120 72

090~099 16 49 7 79 119 61 22 84 9 97

100~109 91 15 21 24 46 39 93 105 65 70

110~113 129 99 17 123

7) Concept synchronous cell

The concept of synchronous cell plays an important role in planning FH strategy and

lowering intra-network interference. BTS and MS achieve synchronization through their

agreement on FN. In synchronous cell, since the FNs of all TRXs are completely the

same, it is possible for different FH groups to use the same HSN. Adjust MAIO to avoid

the collision between cells and the adjacent frequencies of the same cell.

III. Parameter

FH data configuration sequence: CAξMAξHSNξMAIO

Four parameters of FH algorithm: MA = {f1, f2,↑,fN}, HSN, MAIO, FN

The data tables and parameters involved in configuration are detailed below.

[Cell/Modify Cell's FH Property/Modify Cell FH]

Parameter: FH Mode

Parameter: FH Group Assignment Table

[Cell/Modify Cell's FH Property/Modify Cell FH /Configure MA Group]

Parameter: MA Frequencies Assigned

Parameter: Current MA Group No.

Parameter: FH Sequence No.





2-60

Parameter: Training Sequence No. (TSC)

2.2 Extended Functions

2.2.1 Handover

I. Overview

Handover is a very important function in a cellular mobile communication network.

Handover enables the continuous session of subscribers while moving around different

cells. Besides, handover can also adjust the traffic of the cell, thus optimizing the

overall performance.

The overall handover process is implemented in the MS, BSS and MSC. Measurement

of radio subsystem downlink performance and signal levels received from surrounding

cells, is made in the MS. These measurements are send to the BSS for assessment.

The BSS measures the uplink performance for the MS being served and also assesses

the signal level of interference on its idle traffic channels. Initial assessment of the

measurements in conjunction with defined thresholds and handover strategy may be

performed in the BSS. Assessment requiring measurement results from other BTS or

other information resident in the MSC, may be performed in the MSC. In the above

handover process, handover decision algorithm is the most important part because it

determines the service quality and frequency efficiency.


Huawei handover algorithm includes cell sequencing and handover judgement.

1) Cell sequencing

The cell sequencing can be divided into two parts: basic sequencing and network

feature adjustment.

a) Basic cell sorting.

Huawei handover algorithm adopts the M principle and K principle based on level

comparison in stead of L principle based on path loss. With M principle and K principle,

the serving cell and all adjacent cells are sequenced according to their levels to obtain

the standby cell list on the basis of levels.

M principle: check whether the downlink receiving level of the adjacent cell is higher

than the minimum receiving level while taking uplink and downlink balance

compensation. Only the cell with receiving level lower than the minimum receiving cell,

i. e. RxLEV > MS Rx MIN + MAX (0, Pa), can enter the standby cell list. In that formula

Pa = MS TxPWR MAX - P.





2-61

RxLEV is the MS receiving level for this cell.

MS Rx MIN is the minimum receiving level of MS required by the cell.

MS TxPWR MAX is for restricting the maximum transmitting power of MS.

P is the maximum transmitting power of the MS.

K principle:

( ) ( ) ( ) ( )no BIAS K o RxLEV n RxLEV nrank K , _ _ −−=

( ) ( ) ( )( ) KHYST KOFFSET o RxSUFF n RxSUFF n BIAS K −−−= _

After removing KOFFSET (offset) and KHYST (Hysteresis). The formula of K

sequencing:

( ) ( ) ( )( ) ( ) ( )( )o RxSUFF o RxLEV n RxSUFF n RxLEV nrank K −−−= _

In this formula, (RxLEV(n) – RxSUFF(n)) shows the difference between the adjacent

cell receiving level RxLEV (n) and adjacent cell minimum receiving level threshold

RxSUFF (n). (RxLEV(o) – RxSUFF(o)) shows the difference between the serving cell

receiving level RxLEV(o) and serving cell minimum receiving level threshold

RxSUFF(o). These two differences decide the position of an adjacent cell in the

standby cell list.

RxSUFF(n) is adjacent cell minimum receiving level threshold

RxLEV(o) is serving cell receiving level

RxSUFF(o) is serving cell minimum receiving level threshold

Note:

The purpose of hysteresis is to avoid Ping-Pong handover. The communication may be handed over back

and forth due to the unstable signal at the edge of the cells. This causes much increase in the load to the

system. Applying hysteresis is like enlarging the coverage radius of the serving cell while shortening the

coverage of the coverage radius of adjacent cell. In this way, handover will not be easily triggered, and the

Ping-Pong handover can be eliminated.

b) Adjustment according to network features

Network feature adjustment uses the network information except for the power level to

decide the position of each cell in the standby cell list, thus providing the ultimate

standby cell list for handover judgement.

Implemented according to the load of the cell. The cell with less load has the

higher priority.

Implemented according to whether BSC/MSC is the same. The cell controlled by

the same BSC or MSC has the higher priority.





2-62

Implemented according to the layer and level of the cell. The cell of lower layer or

level has the higher priority.

With basic sequencing of cell and network feature structure adjustment, it is possible to

have a best cell list on broad sense. In other words, regardless of the cause that

triggers handover, a cell ranking first in the list is not a result of certain processing

based on certain criteria.

c) Layers and levels of the cell

Hierarchical classification of the network can fulfill the demands of both coverage and

hot spot traffic. This is a mandatory function to be realized. Normally, the macro cell

settles the problem of coverage, while the micro cell tackles the problem of hot spot

traffic.

The basic frame of Huawei network hierarchy has four layers. They are Umbrella,

Macro, Micro and Pico. In the multiband network, the top layer GSM 900 is usually set

as Umbrella, and the major layer of GSM900 is Macro. The major layer of GSM 1800 is

Micro and micro cell of GSM900/GSM1800 is Pico. Besides, it is also possible to

differentiate the priority of GSM900/GSM1800 band according to the cell's layer. There

are 16 levels of priority at each layer. In the network planning following this mechanism,

the network is first considered according to the layers. The lower layer has higher

property. In the same layer, according to the needs of network planning, the GSM 900

and GSM 1800 can be set with different priority. The smaller priority level has the higher

priority.

GSM 900

GSM900 GSM900 GSM900GSM900 GSM900

GSM1800 GSM1800 GSM1800GSM1800 GSM1800

Macro Cell

Pico Cell

Layer 4

Layer 3

Layer 2

Umbrella

Cell

GSM 900 GSM 900 GSM 900

GSM900GSM900

GSM1800GSM1800

GSM900 GSM900

GSM1800 GSM1800

Micro Cell

Layer 1

GSM 1800 GSM 1800 GSM 1800

Figure 2-15 Layers and levels of the cell

2) Operation types

a) TA handover





2-63

TA can be regarded as a criterion for controlling the size of a cell. BSC judges whether

the TA of the current MS exceeds the maximum Timing Advanced LIMit (TALIM). If so, it

will initiate an emergent handover. The value range of TA is 0~63. The step length of

each bit is 553. 5 m, the TA setting can compensate for a distance 35 km over 63 steps.If the serving cell satisfies the requirement of TA handover. After a successful handover,

the original cell will be punished so as to avoid this MS handover back to it for other

causes.

b) BQ handover

The BER values used to define a quality band are the estimated error probabilities

before channel decoding. BSC assesses the quality of radio link according to the

quality level in the measurement report. The correspondence between quality level and

actual BER is shown in Table 2-6.

Table 2-6 BER corresponding to quality level

Quality level BER range Assumed value Calculated value

0 Less than 0. 2% 0. 14% 14

1 0. 2% to 0. 4% 0. 28% 28

2 0. 4% to 0. 8% 0. 57% 57

3 0. 8% to 1. 6% 1. 13% 113

4 1. 6% to 3. 2% 2. 26% 226

5 3. 2% to 6. 4% 4. 53% 453

6 6. 4% to 12. 8% 9. 05% 905

7 Greater than 12. 8% 18. 10% 1,810

The cause of BER increase could be signal power too low or channel interference.

When the receiving quality in the serving cell is lower than the BQ handover threshold,

BQ handover will be started so that the MS can maintain transmission quality of a

certain level. If the serving cell satisfies the requirement of BQ handover. After asuccessful handover, the original cell will be punished so as to avoid this MS hands

over back to it for other causes.

c) Signal level rapid dropping handover

Handovers such as edge handover and PBGT adopt methods such as averaging filter

and P/N judgement. However it is not sensitive to short term signal level rapid dropping,

FIR (Finite Impulse Response) filter to original receiving signal level is used to settle

this problem. This kind of filter has a quick response to the rapid dropping slope of the

original receiving signal level signal.





2-64

Note:

Since the parameter setting of level rapid dropping handover algorithm is rather complicated, it is not easy

to obtain experience point. Therefore, this function is usually not used.

d) Interference handover

When the receiving level in the serving cell is high, but the receiving quality deteriorates

to a certain extent, interference handover is started so that the MS can maintain

transmission quality at the certain level. Difference between interference handover and

bad quality handover: in the former case, the quality is not low enough to affect session,

and the receiving level is still high.

When the active channel quality is affected by little interference in the serving cell, but

they still can sustain the ongoing communication. At the same time, the receiving level

in the serving cell is higher. There is possibility less interference on other channels in

the serving cell, so intra-cell handover can be carried out.

The parameters of interference handover algorithm: Qual_Thr and Lev_Thr. This to

decide whether to trigger interference handover. If RxLev > Lev_Thr and RxQual >

Qual_Thr, the interference handover is triggered. Interference handover is illustrated in

Figure 2-16.

Receivingquality(dtqu)

Receivinglevel

0

Qual_Thr

Lev_Thr

Figure 2-16 Interference handover zone

The shadowed part in the figure stands for zones within which interference handover

occurs.

e) Edge handover

This is a level-based handover and is rescue handover. If edge handover is to be

triggered, the level of the destination cell is required to be higher than that of serving

cell for at least one hysteresis value (inter-cell handover hysteresis).





2-65

The criterion for triggering edge handover: When the receiving level of the serving cell

is lower than the edge handover threshold, and fulfilling the P/N criterion within a

certain measurement period, the edge handover will be triggered to ensure the

communication quality. Edge handover is illustrated in Figure 2-17.

Cell1 Cell2

-97dBm -85dBm

Figure 2-17 Edge handover

f) PBGT handover

PBGT also belongs to better cell handover, a handover based on path fading. PBGT

handover algorithm searches for the cell with lower path loss and satisfying the system

requirement on real-time basis so as to judge whether handover is needed. Difference

from other handover algorithms: the trigger condition is path loss and receiving power.

Triggering condition of PBGT handover: The path loss of the adjacent cell is smaller

than the threshold of the serving cell and the P/N criterion is satisfied within a period of

measurement time. P/N criterion is that there are P satisfying the criterion during N

measurements.

PBGT(n) > PGBT_HO_Margin (n)

In the inequality above, P, N and PBGT_HO_Margin (n) are configured at data

configuration console. PBGT (n) calculates according to the control parameter and the

information reported by BTS.

The method of calculating PBGT (n):

( ) ( )( )

( )( ) ( )( )n NCELL RxLEV P nMAX TxPWRMS Min

DC PWR DL RxLEV P MAX TxPWRMS Minn PBGT

_ , _ _

_ _ _ , _ _

−−

−−=

Meanings of the parameters:

MS TxPWR MAX: maximum transmitting power allowed in the serving cell

MS TxPWR MAX (n): maximum transmitting power allowed in the adjacent cell n

RxLEV_DL: downlink receiving power of the serving cell

RxLEV_NCELL (n): downlink receiving power of the adjacent cell n

PWR_C_D: difference between the maximum downlink transmitting power caused

by power control and the actual downlink transmitting power of the serving cell.





2-66

P: maximum transmitting capability of MS

PBGT handover occurs only between cells of the same layer and same level.

g) Load handover

There may be cells with heavy load while their upper layer cell and the adjacent cell

bears less load. To achieve load balance between cells by sharing the load with upper

layer and adjacent cell, the traffic load handover is applied. Its aim is to hand over part

of the traffic in the heavily loaded cell to less loaded cells, and preventing the traffic of

the adjacent cells being handed over to this cell. Load handover can be implemented

between cells within the same BSC. Load handover is illustrated in Figure 2-18.

Low traffic cell

Low traffic cell

Low traffic cell

Heavy traffic cell

High traffic cell

Low traffic cell

High traffic cell

Figure 2-18 Load handover

The method of realizing load share: by heightening the edge handover threshold

towards that of the serving cell, the traffic at the cell edge will be handed over those with

less traffic. The basis for judging the traffic of a cell is the cell flow (i. e. TCH occupation

rate) and the preset threshold. If the cell flow of a cell is higher than the heavy traffic

load threshold (Load HO Start Threshold ), this cell is consider to have a heavy traffic

load, and the load handover algorithm needs to be activated. If the cell flow of a cell islower than the low traffic threshold (Load HO Rx Threshold), it is consider having a low

traffic load, and is allowed to accept the traffic handed over from other heavy traffic load

cells.

Since the load handover mechanism is likely to trigger a good number of handovers,

the situation of system CPU load should be taken into consideration before triggering

handover, i. e. system flow level. In addition, to avoid too many handovers happening

simultaneously, the load handover is implemented step by step, i. e. edge handover

threshold will increase by certain step length (CLS_Ramp) and period (CLS_Period).





2-67

The increase ends when the threshold reaches the load handover bandwidth

(CLS_Offset). Load handover is illustrated in Figure 2-19.

Cell A Cell B

Normal HO border

CONF_HO_RXLEV

CONF_HO_RXLEV+CLS_Ramp

CONF_HO_RXLEV+CLS_Offset

Load HO zone

Figure 2-19 Load handover

h) Hierarchical handover

The GSM network is classified into layers, so as to flexibly direct its traffic and fulfill the

needs of different network structure.

If a cell has a high priority and its signal level is higher than a threshold (Inter-layer HO

Threshold) and satisfy the P/N criterion, the traffic will be handed over to this cell even

if the serving cell can still provide normal services. The purpose of hierarchical

handover is to direct the traffic to the cell with higher priority so that the traffic can be

distributed more reasonably.

i) Fast moving handover

This kind of handover is carried out for fast moving MS to reduce the number of

handover and hence reduced call drop rate.

If MS is moving quickly with micro cell as the reference, it will be handed over to the

macro cell. If the fast moving MS registered in the macro cell, time penalty will be

implemented to the micro cell so that the MS will stay in the macro cell. Fast moving

handover is illustrated in Figure 2-20.





2-68

Umbrella Cell

Micro Cell

Figure 2-20 Fast moving handover

There are two principles for fast moving handover:

If the MS is moving fast with the micro cell as the reference, it will be handed over

to the macro cell.

To avoid the fast moving MS registered in macro cell enter the micro cell, time

penalty will be implemented to micro cell.

If the duration of MS camping in a cell is lower than a certain threshold (Fast Moving

Time Threshold), this MS is considering to be moving fast with this cell as the reference.

To avoid miscarriage of justice, P/N measurement will be implemented to several cells.

If the criterion of fast moving is satisfied, this MS will be handed over the macro cells.

For MS registered in macro cell, the method of "timer + penalty" is applied. Before the

speed sensitive timer of a certain micro cell times out, this receiving level of this micro

cell will be punished, so that the position of this micro cell in the cell sequencing will be

lowered.

Fast moving handover algorithm can only perform accumulation judgement to the MS

within the same BM and same BSC. When MS moves to another BM, it is necessary to

re-judge.

j) Other handovers

Other handovers include IUO handover, directed retry, forced handover, and extended

cell handover.

3) Handover procedure

Handover decision algorithm enables the preprocessing of the input MR and decides

whether handover should be done and which type of handover it should be (intra-cell

handover, inter-cell handover in the same BSC, outgoing BSC handover, etc.)

according to the various conditions. Handover decision algorithm sends the message

of decided handover result to call process module, which will complete

handover-signaling process together with BTS, MSC and MS. If a handover fails for a





2-69

certain reason, call process module will notify the handover result to the algorithm,

which will decide how to further process this handover. Handover process is as shown

in Figure 2-21.

MR preprocessing

MR averaging procesing

Handover decision

Penalty processing of cell measurement value

Basic cell sorting

Adjustment according to network features

Sending handover commands to the call handling module

Processing of handover results

Handover decision algorithm starting decision

Call control

Figure 2-21 Handover decision process flow chart

Each phase of the process is described as follows.

a) MR preprocessing.

MR provides basic parameters needed in handover decision. MS measures the

receiving quality (RxQual) and receiving level (RxLev) of the downlink of the serving

cell as well as the downlink RxLev of the BCCH carrier frequency of adjacent cells (best

six adjacent cells average). Then MS sends these measurement results to BTS through

SACCH once every 480ms. If SACCH is used for the transmission of other signals, MS

sends the measurement results once every 960ms. BTS measures the RxQual and

RxLev of the corresponding uplink. BTS combines the uplink measurement value and

the downlink measurement value to form a MR message.





2-70

Note:

If messages transmitted on the uplink SACCH do not include the MR (transmitted by MS), the uplink

measurement result will indicate that the MR transmitted by MS is lost.

MR should be preprocessed so that it can have a better reflection of radio links. MR

preprocessing process can be realized in both BTS and BSC and controlled by OMC.

The preprocessing of the MR includes the following three procedures:

MR interpolation processing: When discontinuous MRs are received by BTS or BSC,

lost MRs should be interpolated so as to guarantee the continuity of the whole MR

processing process. This procedure is called MR interpolation. If the number of lost

MRs exceeds a limit, previously received MRs will be regarded as invalid ones and

re-collection is needed.

To eliminate the uncertainty in handover decision, it is necessary to perform smooth

processing over the MRs, or filtering. A simple and practical algorithm is weighted

filtering. Different filter lengths can be respectively defined for different types of

measurement values like the receiving level, receiving quality and TA, or different

channel types like signaling channels, speech and data channels. The receiving level

(RxLev) and receiving quality (RxQual) use corresponding assumed value for

calculations, as shown in Table 2-7, and Table 2-8.

Table 2-7 Receiving level calculation assumed value

RxLev number Implication Assumed value

0 < -110dBm -110dBm

1 -110dBm~109dBm -109dBm

2 -109dBm~-108dBm -108dBm

… … …

62 -49dBm ~ -48dBm -48dBm

63 > -48dBm -47dBm

Table 2-8 Receiving quality calculation assumed value

RxQual number BER range Assumed value Calculated value

0 < 0. 2% 0. 14% 0

1 0. 2% ~ 0. 4% 0. 28% 10

2 0. 4% ~ 0. 8% 0. 57% 20





2-71

RxQual number BER range Assumed value Calculated value

3 0. 8% ~ 1. 6% 1. 13% 30

4 1. 6% ~ 3. 2% 2. 26% 40

5 3. 2% ~ 6. 4% 4. 53% 50

6 6. 4% ~ 12. 8% 9. 05% 60

7 > 12. 8% 18. 10% 70

The MR represents the condition of radio channels in the previous measurement cycle,

so it is of hysteresis to some extent. The prediction algorithm is mainly responsible for

MR values for the next cycle(s) based on the radio environment changes prediction.

MR prediction is a process that can be selected by the operator.

When the multiplexing on the Abis interface is 15:1, every 4 signaling links multiplex a

64kbit/s timeslot statistically. MR is transmitted through RSL. In order to minimize

signaling transmission error bit, when Abis interface multiplexing mode is 15:1, MR

processing mode requires that MR should be a preprocessed one instead of the

original one. Moreover, MR reporting frequency can adopt interval reporting. It can be

realized with data configuration: in [Cell/Modify Cell's Handover parameter/Modify

Handover Parameter/HO Control Data], modify [BTS Measurement Report

Preprocessing], [Transfer Original Measurement Report] and [Report Freq. of

Preprocessed Measurement Report]. Accordingly, the emergency handover due tofast level dropping is decided by BTS. And the BSC will forward the decision. For other

handovers completed within BSC, handover decisions and processing are still carried

out in the BSC.

b) Handover decision algorithm starting decision

Judge whether basic conditions for handover are satisfied, such as whether there are

enough MRs. If conditions are satisfied, handover decision algorithm is started.

c) MR averaging processing

Filter MRs according to a certain algorithm, cancel their noise, and smooth MRs, thus to

prevent incorrect handovers due to individual interference.

d) Penalty processing of cell measurement value

Practically there is a possibility that a handover can not be successful. In case the

handover to the selected target cell fails, the MS will stick to the original serving cell.

After the cycle of a handover decision is finished, the system might try to hand over the

MS to the above-mentioned target cell again, which might cause invalid handover

attempt or handover failure, or even interruption. Therefore, the target cell shall be





2-72

punished, which is to reduce the receiving power of the corresponding cell by a set

penalty value for a period (called the penalty time).

Penalty types include the penalty of the forsaken cell due to TA value, penalty of the

forsaken cell due to bad quality (BQ), penalty to the failed cell due to ordinary handover

failure, and fast-moving penalty (this is a penalty imposed on the microcell in the

candidate queue in order to prevent frequent handover when the fast moving MS

accesses a cell of small coverage).

e) Basic cell sorting.

Adjacent cells that have been imposed penalty and the serving cell are sorted through

a certain algorithm, thus the position of each cell is located. This is to get ready for final

handover.

f) Adjustment according to network features

To adjust candidate queue through a certain algorithm according to hierarchical

network, cell priority, speed sensibility, and the specific network environment.

g) Handover decision

Handover decision algorithm is used to decide the time to start handover and the target

cell to be handed over. Confirm the candidate cell queue list, adjust cells adjusted in

last procedure and finalize a uniform clear list of cells that are ready to be handed over.

h) Sending handover commands to the call handling module

After making the handover decision with the algorithm and deciding to execute the

handover, BSC sends handover message containing the type of incoming handover to

the call-handling module, then the latter starts the signaling procedures for this

handover.

i) Processing of handover results

After the call handling module has processed handover signaling, it returns the result to

the handover decision module. If the handover fails, the handover decision module will

start penalty to the cell responsible for the failure. If the handover is successful, themodule will set a new handover interval timer to avoid frequent handovers.

III. Parameter

1) TA Handover

“TA Thrsh ” in [Handover\Emergency Handover Table]

“Filter Length for TA ” in [Handover\Filter Table]

“Penalty Time after TA HO ” in [Handover\Penalty Table]

“Penalty Level after TA HO ” in [Handover\ Penalty Table]





2-73

2) BQ Handover

“UL Qual. Thrsh ” in [Handover\Emergency Handover Table]

“DL Qual. Thrsh ” in [Handover\ Emergency Handover Table]

“Filter Length for TCH Level ” in [Handover\Filter Table]

“Filter Length for SD Qual ” in [Handover\Filter Table]

“Penalty Level after BQ HO ” in [Handover\Penalty Table]

“Penalty Time after BQ HO ” in [Handover\Penalty Table]

3) Level rapid dropping handover

“Rx_Level_Drop HO Allowed ” in [Handover\Handover Control Table]

“Filter parameter A1~A8 ” in [Handover\Emergency Handover Table]

“Filter parameter B ” in [Handover\Emergency Handover Table]

4) Interference handover

“UL Qual. Thrsh. for Interf. HO ” in [Handover\Emergency Handover Table]

“DL Qual. Thrsh. for Interf. HO ” in [Handover\Emergency Handover Table]

“UL RX_LEV Thrsh. for Interf. HO ” in [Handover\Emergency Handover Table]

“DL RX_LEV Thrsh. for Interf. HO ” in [Handover\Emergency Handover Table]

5) Edge handover

“Edge HO UL RX_LEV Thrsh ” in [Handover\Normal Handover Table]

“Edge HO DL RX_LEV Thrsh ” in [Handover\Normal Handover Table]

“Edge HO watch time” in [Handover\Normal Handover Table]

“Edge HO valid time” in [Handover\Normal Handover Table]

“Inter-cell HO Hysteresis” in [Handover\Adjacent Cell Relation Table]

6) PBGT handover

“PBGT HO Allowed” in [Handover\Handover Control Table]

“PBGT HO Thrsh” in [Handover\Adjacent Cell Relation Table]

“PBGT Watch Time” in [Handover\Normal Handover Table]

“PBGT Valid Time” in [Handover\Normal Handover Table]

7) Load handover

“load HO Allowed ” in [Handover\Handover Control Table]

“System Flux Thrsh. for Load HO” in [Handover\Load Handover Table]





2-74

“Load HO Thrsh” in [Handover\Load Handover Table]

“Load Req. on Candiate Cell” in [Handover\Load Handover Table]

“Load HO Bandwidth” in [Handover\Load Handover Table]

“Load HO Step Period” in [Handover\Load Handover Table]

“Load HO Step Level” in [Handover\Load Handover Table]

8) Layered and hierarchical handover.

“Layer of The Cell” in [Handover\Cell Description Table]

“Cell Priority” in [Handover\Cell Description Table]

“Inter-layer HO Thrsh” in [Handover\Cell Description Table]

“Inter-layer HO hysteresis” in [Handover\Cell Description Table]

“Layer HO watch time” in [Handover\Normal Handover Table]

“Layer HO valid time” in [Handover\Normal Handover Table]

9) Fast Moving handover

“MS Fast Moving HO Allowed” in [Handover\ Handover Control Table]

“MS Fast-moving Watch cells” in [Handover\Fast-moving handover Table]

“MS Fast-moving Valid cells” in [Handover\Fast-moving handover Table]

“MS Fast-moving time Thrsh” in [Handover\Fast-moving handover Table]

“Penalty on MS Fast Moving HO” in [Handover\Cell Description Table]

“Penalty Time on MS Fast moving HO” in [Handover\Cell Description Table]

2.2.2 Power Control

I. Overview

As an important method to control radio link, power control adjusts the transmit power

of MS and BTS according to the expected value configured in OMC data management

system, the receiving level (including uplink and downlink) from BTS and the MR of

receiving quality. Basic rules for power control are:

1) When the level or signal quality is higher than the expected value, the power

should be decreased accordingly.

2) When the level or signal quality is lower than the expected value, the power should

be increased accordingly.

3) The level and signal quality should be both considered so as to improve the

accuracy and effectiveness.





2-75

The nature of a cellular system requires that the output power of the BSC and MS

should be set as low as possible. With the limited resource of the RF spectrum cellular

systems depend upon the reuse of the RF channels. The reuse distance between

these channels mainly upon the subscriber density in a particular area, the greater thedensity the shorter the reuse distance. By keeping the MS and BSC at the minimum

acceptable power output it reduces the chances of interference, particularly

co-channel.

Another benefit of effective power control is that the battery of MS is extended, thus

maximizing available talking time.

Huawei BSS offers three different algorithms for the implementation of power control,

which is GSM 0508 power control algorithm, and Huawei I (HW_I) and Huawei II

(HW_II) algorithms. Any algorithm can be selected among these three algorithms.

HW_I and HW_II algorithms are recommended due to their flexible configurations,

effectiveness, easy operations and easy command. These Huawei-developed

algorithms are compatible nicely with the GSM900 and GSM1800 systems.


1) Power control classification

Power control comprises uplink and downlink power controls, which are executed

separately. The uplink power control is for MS while the downlink power control is for

BTS.

a) MS power control

The purpose of MS power control is to adjust the MS output power in order to achieve

the stable receiving signal so as to reduce the interference from subscribers of adjacent

channels, decrease the saturation degree of BTS multicoupler and reduce MS power

consumption.

The MS power control is divided into two adjusting stages, i.e., the stable adjusting

stage and the initial adjusting stage. The stable adjusting is the normal method for

performing the power control algorithm, while the initial adjusting is used in the time

when the call connection is initially started. When a connection is performed, MS is

output as the nominal power of the cell where it is located (the nominal power indicates

that the MS transmitting power is the MS maximum transmitting power

MS_TXPWR_MAX_CCH in the broadcast system messages on the BCCH channel of

the cell where it is located. If MS does not support this power class, the supported

power class that is nearest to it will be utilized, such as the maximum output power

class supported by the reported MS Classmark in the establishment indication

message). However, since BTS may simultaneously support multiple calls, the

receiving signal intensity should be reduced in a new connection as quick as possible,

otherwise, the quality of other call supported by this BTS may be deteriorated due to





2-76

the saturation of the BTS multi-coupler, and the call quality of other cells may be

affected due to the high interference. Therefore, the purpose of the initial stage power

control adjusting is to reduce the MS transmitting power as quick as possible until the

stable measurement report is obtained, so that the MS can be adjusted according to thestable power control algorithm.

The parameters that must be selected in the uplink power control, such as the expected

desirable uplink receiving level, desirable uplink receiving quality, etc. , are all set by the

O&M data management console, the data configuration can be dynamically carried out

according to the actual situations of the cell. After a given number of the uplink

measurement reports is received, by the processing methods such as interpolation and

filtering, the actual uplink receiving level and the receiving quality are obtained, then

they are compared with the desirable uplink receiving level and the receiving quality,

with the power control algorithm, the power class to which the MS should be adjusted iscalculated; if it is different from the current MS output power class and meets a given

application restricted conditions (such as the power adjusting step length restriction,

MS output power range restriction), the power adjusting command is sent. The

essence of the uplink power control adjusting is to enable the actual uplink receiving

level and receiving quality obtained from interpolation and filtering to progressively

approach the desirable uplink receiving level and receiving quality set by O&M. The

purpose for the interpolation and filtering of the measurement reports is to process the

lost measurement report, clear the temporary nature (spiffiness), so as to ensure the

stability of the power control algorithm.

The difference between power controls in initial phase and stable phase is that their

expected uplink receiving levels and receiving qualities, filter lengths are different, and

the former one only adjusts downwards.

b) BTS power control

The BTS power control is an optional function. The base station power control is

basically identical to the MS power control, except that the base power control utilizes

only the stable power control algorithm. The parameters that must be selected in the

power control include the receiving level threshold (lower limitation) to be performed the

power control and the receivable maximum sending level threshold (upper limitation).

The receiving level RXLEV is divided into 64 classes, with numbers from 0 to 63, class

0 of the receiving level is the lowest, while the class 63 of the receiving level is the

highest. The base station power control is divided into the static power control and the

dynamic power control, the later is the fine adjusting based on the former. The GSM

05.05 protocol specification specifies that the base station static power class is divided

into 6 (2dB/per class), when the maximum power output by the base station is 46dBm

(40W), the class 6 is 34 dBm. The static power level is defined in the cell attribute table

of the data management console, i. e., the maximum output power value Pn of the

current dynamic power control is specified. As the dynamic power control classes are





2-77

set to 15, the range of the dynamic power control is Pn-Pn-30dB. If the requirements

cannot be satisfied when the dynamic power control reaches its maximum value, the

static power control classes should be adjusted to increase the maximum output power

value Pn of the dynamic power control. enable the actual uplink receiving level andreceiving quality obtained from interpolation and filtering to progressively approach the

desirable uplink receiving level and receiving quality set by O&M. The purpose for the

interpolation and filtering of the measurement reports is to process the lost

measurement report, clear the temporary nature (spilliness), so as to ensure the

stability of the power control algorithm.

2) Execution process of power control

There are 3 MR cycles from sending command to execution, as shown in Figure 2-22.

SA0 SA1SA0 SA0SA1SA1 SA2SA2SA2 SA3SA3SA3

BTS transmifs the command

of adjust power and TA at

SACCH header

MS obtains SACCH block

MS starts to send the

messurement report of

the previous multi-frame

In the 26 multiframethe 12th frame is for

sending SACCH

BTS receives the

measurement report

Report period of SACCH:

26× 4 104 frame (480ms)＝

MS adopts new

powerand TA

MS Generates new SACCH

header to report new TA and

power control message

Figure 2-22 Power control execution process

a) In the first MR cycle, MS receives the power regulation message carried by SACCH

header on dedicated channel and the first layer header carried by a downlink SACCH

message block. MS will execute the power control command in next cycle instead of

upon the receipt of these headers in first cycle.

b) In the second MR cycle, power control is executed. The maximum rate of change of

MS power is 2dB/13 frame (60ms). If the regulation step length is 8, i. e. 8×2=16dB. It

needs 104 frames (i. e. 480ms, one MR cycle) to complete power regulation. If the

regulation step length is 16, i. e. 16×2=32dB. It takes 2 MR cycleS to complete power

regulation.

c) In the third cycle, the current transmit power (refers to the power level used by the

last burst pulse of SACCH MS cycle) is stored, which will be reported to BTS in next

SACCH uplink MR.





2-78

3) Power Control Algorithm

BSC can dynamically implement power control on each MS and BTS Three algorithms

can be adopted as power control algorithm: GSM 0508 algorithm, HW_I algorithm and

HW_II algorithm. Algorithm process is as shown in Figure 2-23

Power Control Algorithm selection

HW_I Power control algorithmGSM0508 power control algorithm

MR Preprocessing

HW_II Power control algorithm

Figure 2-23 Power control algorithm selection

Power control algorithm is specified in the 0508 protocols (for further details refer to the

related specifications). For uplink, upper limit and lower limit thresholds are set for the

receiving signal level and the receiving signal quality. Counters P' and N' are used to

count the MR and the values of these counters can be set through OMC. When N' MRs

in the consecutively received P' MRs exceeds the above threshold, power regulation

will be executed.

Usually the steps for power control are:

MR preprocessing

Power calculation

Power control decision

Adjustment by sending power control commands

4) Huawei HW_I algorithm

Huawei HW_I has following features:

Compared with protocol algorithm, the initial state regulation is added.

Data configuration is rather complicated. The power control adjustment involves

many parameters and complicated calculation.

Power control decision is the sum of the level and quality, and the expected value

is just a specified value instead of a range. Once the adjustment results of the

receiving level and receiving quality are contrary, the power control will never stop

and the level fluctuates with the expected value.

Power control decision process

HW_I algorithm power control decision process is as shown in Figure 2-24.





2-79

MR pre-processing

Satisfying power control target

Y

N

Power control calculation

and regulation cinitial

state and stable state

Figure 2-24 HW_I algorithm power control decision process

b) Measurement Report

In order to implement power control decision, various kinds of information about the

current communications status from MS and BTS should be collected, including

receiving signal level, and communication quality etc. Network side on SACCH will

receive MRs from MS and BTS every 480 ms, in which various kinds of information

needed for power control decision are contained. The process of MS reporting is as

shown in Figure 2-25.

MR

MR

MRMR

Downlinkmeasurement

Uplinkmeasurement

Figure 2-25 Reporting MR

An example of BSS MR is shown in Figure 2-26.





2-80

Figure 2-26 MR example

c) MR preprocessing.

Interpolation: each MR has a serial number. If the serial numbers of received MRs are

found not continuous, this means that there are some MRs gets lost. In this case,

network will add all lost MRs according to interpolation algorithm.

Filtering: Several continuous MRs results will be used to reflect the state of MS in a

period of time thus to avoid the one-sidedness caused by judging the state of MS

according to only one MR result.

d) Power control decision

Number of transmit power to be adjusted

(Expected stable signal level - current receiving signaling level) × uplink (downlink)

compensating factor + (current actual receiving uplink (downlink) quality – expected

uplink (downlink) quality) × 10 × uplink (downlink) quality compensating factor

Caution:

The last regulated power level cannot exceed the maximum power control step length.





2-81

Actual stable level equals to the sum of current actual level and transmit power to be

regulated

During the process of judging power control level to be adjusted, it needs to search

tolerance table according to the level of current transmit power. If the final power

regulation level is with the tolerance range, the regulation is unnecessary. GSM1800

tolerance table is shown in Table 2-9.

Table 2-9 GSM1800 tolerance table

Level 0 1 2 3 4 5 6 7 8 910

11

12

13

14

15

16

17

18

19

Tolerance 2 2 2 2 2 2 2 2 2 3 3 3 3 3 4 4 4 2 2 2

GSM900 tolerance table is shown in Table 2-10.

Table 2-10 GSM900 tolerance table

Level 0 1 2 3 4 5 6 7 8 910

11

12

13

14

15

16

17

18

19

Tolerance 2 2 2 4 4 4 4 4 4 4 4 4 4 4 4 4 6 6 6 6

The similarities and difference of HW_I algorithm uplink power control and downlink

power control is as follows:

Similarities:

In order to avoid the fluctuation caused by power controls, the interval between

two continuous controls is specified for both uplink and downlink.

In order to not being affected by unexpected factors, all MRs should be filtered.

Both uplink and downlink power controls have power control on level and quality

respectively.

Both have maximum power control step length and compensating factor.

Difference:

MS has power control not only for stable state but also for initial connecting phase

before a call is connected. The purpose is to lower MS transmit power as soon as

possible.

Uplink has measures to improve transmit power in the case of MS handover

failure.

Downlink has the restriction for both maximum and minimum MS transmit power.

5) Huawei HW_II algorithm

Compared with HW_I, HW_II has following advantages:

MR compensation, which makes the power control decision more accurate.





2-82

MR prediction, which reduces power control delay.

Adaptive power control, which sufficiently guarantees the algorithm stability and

high efficiency.

Keep power control target within the range between upper limit and lower limit so

as to avoid power control fluctuation.

Easy and flexible data configuration, which guarantees effective regulation of

network optimized parameters.

Configure the upward and the downward power control step respectively.

a) Power control decision process

HW_II power control decision process is as shown in Figure 2-27.

MR pre-processing

Power control requested

by receiving level

Power control requested

by receicing quality

Conprehensive decision

of power control

Figure 2-27 HW_II power control decision process

b) Request power control according to level

After the preprocessing of MR, power control module compares the current

receiving level with expected value.

Then the transmit level step length is calculated. The regulation is to make the

receiving level closer to the expected value.

When receiving level regulates transmit power, variable step length can be

adopted so that the quick power control can be obtained.

c) Request power control according to receiving quality

After the preprocessing of MR, power control module compares the evaluation value of

the current receiving quality with expected value.

Calculate the transmit step length to be regulated

Improving transmitting power for low receiving quality.





2-83

Decreasing transmitting power for high receiving quality.

Do not adjust the transmitting power when the receiving quality falls between the

upper/lower thresholds.

d) Comprehensive decision of power control

Comprehensive decision of power control is shown in Table 2-11.

Table 2-11 Table of comprehensive decision of power control

Receiving level power controlregulation

Receiving quality power control regulation

Comprehensive decision of power control

ο AdjStep_Lev ο AdjStep_QulοMAX(AdjStep_Lev,

AdjStep_Qul)

ο AdjStep_Lev μ AdjStep_Qul No action

ο AdjStep_Lev No action ο AdjStep_Lev

μ AdjStep_Lev ο AdjStep_Qul μ AdjStep_Lev

μ AdjStep_Lev μ AdjStep_QulμMAX(AdjStep_Lev,

AdjStep_Qul)

μ AdjStep_Lev No action μ AdjStep_Lev

No action ο AdjStep_Qul ο AdjStep_Qul

No action μ AdjStep_Qul μ AdjStep_Qul

No action No action No action

e) MR compensation

Power control module will extract the receiving level and receiving quality of some

history MRs when it implements power control decision. These MRs might be obtained

in different transmit powers. In order to guarantee the accuracy of receiving level to be

used, if the transmit powers in these MRs are different, the receiving level value of

history MRs should be compensated. The interpolated and compensated MRs are

filtered so as to make control power decision more effective.

f) Predict filtering

The power control is a process of transmitting power control based upon the current

received level and the receiving quality. The sending and transmission of power control

command and power adjustment will take certain period of time, so there will exist

certain hysteresis between the receiving change and corresponding transmitting power

adjustment. Filtering prediction enables MR on which power control decision is based

to get closer to the state of power regulation so as to erase delay effectively.





2-84

MR filtering prediction is implemented in a very short time and changes of receiving

level and quality are likely to be continuous. N MRs before the current time are used for

weighted filtering, then 0~3 MRs of after the current time are predicted. Generally, there

are 3 MRs between power control decision and power regulation, which is about 1.5second. As a consequence, the accuracy of prediction is guaranteed. Power control

decision is made after the filtering of predicted MRs, interpolated MRs, and

compensated history MRs.

g) Dual threshold power control algorithm

Dual threshold power control algorithm adopts the following three strategies:

Adjust power control step length according to receiving level: The final purpose of

power control is to obtain the best communication quality at the lowest level. However,

due to the instability of radio link and the external interference, transmit power cannotbe lowered greatly. Therefore, HW_II adopts the power control strategy of dual

threshold so as to try to keep receiving within two thresholds.

Adjust power step length according receiving quality: Generally, the change of

receiving quality is associated with interference. The main interference of GSM comes

to same frequency interference generated from frequency multiplexing. This

interference is interactive. One call increases its power means that it exerts a stronger

interference on the other call. Therefore, the power regulation caused by the change of

receiving quality should avoid the group effect of increasing transmit power due to bad

quality. Receiving quality threshold is also set with dual thresholds. Receiving qualitywith the range between two thresholds needs not to adjust transmit power. While

receiving quality beyond the range should be adjusted. For the power regulation

caused by quality factor should use fixed step length to avoid.

Considering both power control strategies of receiving level and receiving quality

regulation. Considering the requirements of both level and quality. On one hand, both

requirements should be satisfied as much as possible; on the other hand, in the case

that the requirements are not consistent or completely contrary, the stability should be

fully considered to prohibit the unstable regulation process. Therefore, the effect on

power control caused by level and quality should be both considered.

h) Variable step length power control

When variable step length regulation is adopted, if that the level or quality is greatly

different from its expected value, use the larger step length to quickly adjust power; in

the case that the level or quality is slightly different from its expected value, use the

smaller step length to adjust power. Thus, quick and accurate power regulation is

achieved.

i) Adaptive power control





2-85

Adaptive power control is to change power control strategy according different

communication environments. This fact leads to a more effective and more stable

power control. This is reflected in following two aspects:

Power control adjustable maximum step length can be adjusted automatically

according to the different communication environments.

The different power control strategies are adopted for different communication

environments.

j) Adjustment of upper threshold of signal strength

Double-threshold power control algorithm is adopted for power control. For level, there

are upper threshold of uplink (downlink) signal strength and lower threshold of uplink

(downlink) signal strength. When the receiving quality is rather poor, the value of upper

threshold can be increased furthermore. When the receiving quality is good, the lower

value of upper threshold can be adopted so as to reduce the transmit power of mobile

phone or base station. When the receiving quality is rather poor, the higher value of

upper threshold can be adopted so as to improve the communication quality.

k) Configure the upward and the downward power control step respectively System configure the upward and the downward power control step respectively, this

enable the system can control the power rapidly and flexibly according to the actual

network. When the up/down link signal quality or receiving quality become worse

suddenly, system increase the power rapidly to avoid call drop.

III. Parameter

1) HW_I algorithm parameters

“Initial RX_LEV Expected ” in [Power\MS Power Control Table]

“Stable RX_LEV Expected ” in [Power\MS Power Control Table]

“Uplink RX_LEV Compensation ” in [Power\MS Power Control Table]

“UL Qual. Expected ” in [Power\MS Power Control Table]

“UL Qual. Compensation ” in [Power\MS Power Control Table]

“Max PC Step ” in [Power\MS Power Control Table]

2) HW_II algorithm parameters

“Filter Length for UL RX_LEV ” in [Power\HWII Power Control Table]

“Filter Length for DL RX_LEV ” in [Power\HWII Power Control Table]

“Filter Length for UL Qual ” in [Power\HWII Power Control Table]

“Filter Length for DL Qual ” in [Power\HWII Power Control Table]

“MR Compensation Allowed ” in [Power\HWII Power Control Table]





2-86

“UL M.R. Number Predicted ” in [Power\HWII Power Control Table]

“DL M.R. Number Predicted ” in [Power\HWII Power Control Table]

“PC Interval ” in [Power\HWII Power Control Table]

“UL RX_LEV Upper Thrsh ” in [Power\HWII Power Control Table]

“UL RX_LEV Lower Thrsh ” in [Power\HWII Power Control Table]

“UL Qual.Upper Thrsh ” in [Power\HWII Power Control Table]

“UL Qual. Lower Thrsh ” in [Power\HWII Power Control Table]

“DL RX_LEV Upper Thrsh ” in [Power\HWII Power Control Table]

“DL RX_LEV Lower Thrsh ” in [Power\HWII Power Control Table]

“DL Qual. Upper Thrsh ” in [Power\HWII Power Control Table]

“DL Qual. Lower Thrsh ” in [Power\HWII Power Control Table]

“MAX Down Adj. value Qual. zone 0” in [Power\HWII Power Control Table]



“MAX Down Adj. PC Value by Qual.” in [Power\HWII Power Control Table]

"MAX Up Adj. PC Value by RX_LEV" in [Power\HWII Power Control Table]

"MAX Up Adj. PC Value by Qual." in [Power\HWII Power Control Table]

“UL Qual. Bad TrigThrsh ” in [Power\HWII Power Control Table]

“UL Qual. Bad UpLEVDiff ” in [Power\HWII Power Control Table]

“DL Qual. Bad TrigThrsh ” in [Power\HWII Power Control Table]

“DL Qual. Bad UpLEVDiff ” in [Power\HWII Power Control Table]

BTS PC class

2.2.3 Extended Cell

I. Overview

In GSM specifications, the TA of cell has a restriction of 63 bit at the radio interface,

which results that the cell coverage radius should be within 35km. In regions such as

vast land, with scattered subscribers, with low traffic, and the infrastructure facilities

such as transmission and power supply are hard to be constructed or unavailable, the

cell with radius over 35km should be provided. The extended cell breaks the restriction





2-87

of radius within 35km. Supported by BTS hardware, it can cover a range with radius of

120km under its ideal state. Carriers can use this technology to fast construct their

GSM networks with fewer stations and at lower cost, and to attract the mobile

subscribers in special regions so as to improve profit.


When the cell coverage radius exceeds 35km, signal delay will exceed the duration

corresponding with the maximum value 63 bit specified in GSM specifications. If an MS

reaches the ordinary coverage verge, it will transmit at the maximum TA allowed by

system; if the MS continues to move outside of cell range, the system is no longer able

to implement adaptive regulation on TA value due to the TA has reached its maximum,

and part of signaling transmitted by MS will reach BTS receiver at next time slot. It is

this principle that extended cell uses to realize the cell extension, i. e. two continuoustime slots in BTS are specified for each MS call, and the receiving window of BTS

receiver is also extended to a width of two time slots thus the cell coverage radius is

extended to over 35km. In order to enable MSs in extended range to initiate call at any

time, two time slots should be always distributed to BCCH, CCCH and SDCCH.

The frame TDMA of GSM radio interface is composed of 8 time slots. Each time slot is

a channel. Normally, the system uses TA to make the uplink signals of MSs with

different distances reach within the corresponding local time slot. TA supports a

maximum of 63 bit. In order to support the extended MS signals over 63 bit, dual time

slot solution binds odd and even time slots and regards each TDMA frame as only withfour channels: 0/1, 2/3, 4/5, 6/7. . For MS, only channel 0, 2, 4, and 6 are distributed.

The MS in the range 0~35km, its TA value changes within the range 0~63. The TA value

of MS with radius over 35km is always maintained as 63. While BTS demodulates

uplink data in two continuous time slots. TA value of TA in MS has a maximum of

63+156. 25 = 219. 25 bit. The principle of extended cell delay regulation is as shown in

Figure 2-28.





2-88

DELAY<=63

After TA adjustment

After TA adjustment

TS0

TS0

TS1

TS1

TS2

TS2

unlink data

delay>63

demodulation range

Dual times lot extendend cell

Figure 2-28 Principle of extended cell delay regulation

If all carrier frequencies in a cell are set as ordinary ones, this is called cell level dual

time slot solution. If part of carrier frequencies in a cell are set as ordinary ones and

other carrier frequencies are configured as dual time slot ones, and BCCH is located in

dual time slot carrier, then this is called carrier level dual time slot solution.

When carrier level dual time slot extended cell is adopted, there are ordinary carrier

and dual time slot ones. BCCH in dual time slot guarantees the random access of any

areas. The calls within TA value accessed randomly being within 35km radius are

distributed to ordinary carrier; while the calls within 34~120km radius and the incoming

handovers are distributed to dual time slot carrier. For the incoming handovers to be

found as 0~35km ones, the system can handover them again to ordinary carrier. When

the calling MS crosses 35 km line, this will lead to an intra-cell handover, which is from

the dual time slot frequency to the ordinary one or from the ordinary to dual time slot

frequency. The conversion of carrier frequencies between ordinary one and dual time

slot one can be set through BSC data configuration

III. Parameter

“Cell Extension Type ” in [Cell\Cell Attribute Table]

“CH Type ” in [Local Office\Radio Channel Configuration Table]

“TA Thrsh ” in [Handover\Emergency Handover Table]

“TA Thrsh ” in [Handover\Concentric cell Handover Table]





2-89

“TA Hysteresis ” in [Handover\ Concentric cell Handover Table]

2.2.4 IUO

I. Overview

With the development of GSM network, the number of subscribers increases gradually,

so the contradict between short frequency resource and great demand is particularly

obvious. In order to increase capacity, the technology of aggressive frequency reuse

should be used to improve the frequency utilization. However, the aggressive

frequency reuse increases the radio interference greatly and even to affect the

communication quality seriously. Under the circumstance of aggressive frequency

reuse, the IUO technology can be used to avoid or decrease radio interference so as to

guarantee communication quality. The IUO technology divides an ordinary cell into two

service layers: OverLaid subcell and UnderLaid subcell. For the MS in the UnderLaid

subcell, try to distribute the less reuse frequency, such as BCCH frequency; for the MS

in the OverLaid subcell, try to distribute the more reuse frequency, such as frequency

except BCCH. The frequency inside the OverLaid subcell adopts aggressive frequency

reuse mode, which can improve system capacity effectively.


IUO refers to the different carrier circle cells formed by different carrier frequencies in a

cell with difference on coverage. Logically, OverLaid subcell and UnderLaid subcell can

be regarded as two cells because their coverage areas are different, The OverLaid

subcell is the main traffic carrier layer because it has many channels. Its function is to

absorb the most subscribers within the cell coverage area. UnderLaid subcell solve the

problem of coverage and provide service for the areas not covered by overlaid cell. e

The technical description of IUO is as shown in Figure 2-29.

Cell A Cell BUnderLaid subcell

InterferenceSignal

OverLaid subcell

Figure 2-29 Aggressive Frequency Reuse of IUO cell





2-90

As shown above, the IUO divides the cell coverage into OverLaid subcell and

UnderLaid subcell. The carrier frequencies of OverLaid subcell and UnderLaid subcell

can adopt different multiplexing modes. For the OverLaid subcell cell, it adopts more

reuse frequency mode such as 1x3 due to its small coverage. For the UnderLaidsubcell cell, it adopts less reuse frequency mode such as 4x3. After the IUO technology

is employed, compared with Multiple Reuse Pattern (MRP), it can greatly increase the

network capacity and guarantee the network quality because the OverLaid subcell

employs of aggressive frequency reuse mode. In some special cases, the UnderLaid

subcell is configured with only one carrier BCCH with the multiplexing mode of 4x3

being adopted and the rest TCH carrier frequencies are configured in OverLaid subcell

with the multiplexing mode of 1x3 being adopted, then the IUO cell is completely the

same as the cell with the multiplexing mode of 1x3 adopted and the average frequency

multiplexing ratio is the same as that of 1x3 multiplexing. Therefore, in this case, the

IUO can effectively reduce the interference for the whole network and obtain the better

network quality than 1x3 multiplexing without the decrease of network capacity.

The wider coverage can be realized through having the carrier in which BCCH is used

large power amplifier. The power that provided by BCCH carrier is greater than other

carriers, so the coverage distance of different carrier is different. While the cell

coverage area depends on the carrier of smaller coverage, so the coverage area is

greatly restricted. When the IUO technology is employed, the carrier with wide

coverage can be used to serve as UnderLaid subcell to realize the far end coverage of

site; while the carrier with small coverage can be used to serve as OverLaid subcell to

increase the near end capacity of site. In this way, the cell coverage area can be

increased.

Underlaid

Overlaid

Figure 2-30 IUO coverage

After the employment of the IUO cell, the cell coverage area can be greatly increased.

The theoretically added coverage of various typical stations with different combining

modes is shown in Table 2-12.





2-91

Table 2-12 Coverage changes for typical sites after the employment of IUO cell

Number of cellcarrier

frequencies

Combining

mode

Loss of low

loss carrier

Loss of high

loss carrier

Added coveragearea after the

employment of IUO

3 CDU+CDU 1. 0dB 4. 5dB 27%

4,5 CDU+CDU+SCU 1. 0dB 8. 0dB 60%

4,5CDU+CDU+CDU

1. 0dB 4. 5dB 27%

5,6 CDU+CDU+SCU 4. 5dB 8. 0dB 27%

The division of OverLaid subcell and UnderLaid subcell is based on the MS downlinkreceiving level, downlink receiving quality and TA. The division of OverLaid subcell and

UnderLaid subcell of common IUO is based on the "RX-LEV Thrsh." and "RX-LEV

Hysteresis". The division of OverLaid subcell and UnderLaid subcell of enhanced IUO

is based on the "U to O HO received level Thrsh." and "O to U HO received level

Thrsh.". as shown in Figure 2-31.

Figure 2-31 Division of OverLaid subcell and UnderLaid subcell in a common IUO cell





2-92

Note:

The division foundation of OverLaid subcell and UnderLaid subcell is as follows:

OverLaid subcell:Receiving Level>= RX_LEV Thrsh. + RX_LEV Hysteresis and

TA<TA Thrsh – TA Hysteresis and Receiving Quality < Receiving Quality Thrsh.

UnderLaid subsell:

Receiving Level < RX_LEV Thrsh. – RX_LEV Hysteresis or TA >=TA Thrsh + TA Hysteresis or

Receiving Quality >= Receiving Quality Thrsh.

RX_LEV Thrsh., Receiving Quality Thrsh. and TA Thrsh. can be adjusted through data

configuration. Therefore, under the precondition of without affecting the networkperformance indexes, the boarders of UnderLaid subcell and OverLaid subcell can be

adjusted flexibly to let OverLaid subcell and UnderLaid subcell rationally share the

traffic.

Receiving Quality Threshold

U to O HO received level Thrsh.

O to U HO received level Thrsh.

Underlaid subcell

Overlaid subcellTA Threshold

TA Hysteresis

Figure 2-32 Division of OverLaid subcell and UnderLaid subcell in a enhanced IUO cell





2-93

Note:

The division foundation of OverLaid subcell and UnderLaid subcell is as follows:

OverLaid subcell:Receiving Level>= U to O HO received level Thrsh. and

TA<TA Thrsh – TA Hysteresis and Receiving Quality < Receiving Quality Thrsh.

UnderLaid subsell:

Receiving Level < O to U HO received level Thrsh. or TA >=TA Thrsh + TA Hysteresis or

Receiving Quality >= Receiving Quality Thrsh.

1) Channel assignment technology of IUO cell

This technology can adopt different assignment strategies in various channelassignment cases in fully consideration of features of IUO. The following are the main

cases:

a) Immediate assignment

System assigns channel through access_delay in Channel Request message. System

assigns the channel in overlaid subcell for the MS in the overlaid subcell; system

assigns the channel in underlaid subcell for the MS in the underlaid subcell. System

always selects the appropriate service layer for MS.

There is no reference receiving level, receiving quality and TA for immediateassignment. In order to guarantee the service quality, the SDCCH of UnderLaid subcell

is assigned preferentially. Only when there is no signaling channel available in the

UnderLaid subcell, will the signaling channel in the OverLaid subcell be assigned.

b) Assignment

The channel assignment strategy of IUO is used to assign channels. The OverLaid

subcell channel will be assigned as far as possible when the subscriber is in the

OverLaid subcell coverage. The UnderLaid subcell channel will be assigned when no

OverLaid subcell channel is available. Similarly, the UnderLaid subcell channel will be

assigned as far as possible when the subscriber is in the UnderLaid subcell coverage.

The OverLaid subcell channel will be assigned when no UnderLaid subcell channel is

available. Select the suitable service layer to serve the subscriber.

c) Intra-BSC handover

Intra-BSC handover is applicable to the non-IUO handover and the handover from the

OverLaid subcell directly to an adjacent cell. Use the IUO channel assignment strategy

to assign channels and select the suitable service layer to serve the MS.

d) Inter-BSC handover





2-94

Being unable to get the receiving level, receiving quality and TA of adjacent cells, the

system selects the preferential UnderLaid subcell, or preferential OverLaid subcell, or

non-strategy mode through switch.

2) IUO cell handover technology

Huawei handover algorithm has the IUO handover judgement function to realize the

IUO technology. When the MS crosses the boundary between OverLaid subcell and

UnderLaid subcell, the IUO handover can be initiated to enable the MS to setup a call at

a suitable service layer. If the object handover layer is congested, the handover will not

be initiated. With the IUO cell handover technology, BSC can intelligently direct the

traffic so as to utilize the frequency resource effectively.

III. Parameter

Parameters in [Handover/ Concentric Cell Handover Table]:

"Direction for IUO HO – UL to OL HO Allowed"

"Direction for IUO HO – OL to UL HO Allowed"

"Criterion for IUO HO – Rx_Lev for UO HO Allowed"

"Criterion for IUO HO – Rx_Qual for UO HO Allowed"

"Criterion for IUO HO – TA UO HO Allowed"

"UO signal intensity difference "

"RX_LEV Thrsh."

"RX_LEV Hysteresis"

"Receiving Quality Thrsh."

"TA Thrsh."

"TA Hysteresis"

"IUO HO Watch Time"

"IUO HO Valid Time"

"Assign optimum layer"

"Assign-optimum-level thrsh."

"Assign-Optimum-TA Thrsh"

"TA pref. Of Imme-Assign Allowed"

"TA Thrsh. Of Imme- Assign pref."

"Incoming-to-BSC HO optimum layer"





2-95

"Pref. subcell in HO of intra-BSC "

"Enhanced IUO allowed"

"O to U HO received level Thrsh."

"U to O HO received level Thrsh."

"U to O Traffic HO Allowed"

"Traffic Thrsh. of underlay"

"Underlay HO Step period"

"Underlay HO Step level"

"Penalty Time of U to O HO (S)"

Parameters in [Handover/Penalty Data Table]:

"Penalty time after IUO HO Fail."

Parameters in [Handover/Cell Description Data]:

"Cell Type"

Parameters in [Site/Carrier Configuration Table]:

2.2.5 "HW-IUO Property"Satellite Transfer

I. Technical description

Satellite communication is the development and the special form of microwave

communication, the supplement and backup to conventional communication means.

Satellite communication features wide coverage, little effected by landform, fine

mobility, and flexible link calling. Meanwhile, it has the problems such as delay, jitter,

and bit error, which leads to the Abis interface of ordinary GSM equipment not

supporting satellite transfer.

Huawei BSS adopts dedicated satellite transfer equipment to realize the satellite

transfer of Abis interface according to the features of satellite transfer. The solution

principle is described as follows:

1) LAPD protocol processing

During the LAPD protocol process, the timer duration is prolonged and the value of

slide window is increased to resist delay.

2) TRAU frame algorithm

The adjustment algorithm of the TRAU frame is modified from fixed cycle adjustment to

self-adaptive adjustment.





2-96

3) BTS clock work mode

The transmission between BSC and BTS can only occupy 19 time slots of DDN circuit

(TS1~18, TS31) and the time slot 0 of DDN circuit is used for the synchronization of

DDN instead of transmitting service. Therefore, BTS can only use the clock of DDN.

However, the accuracy of DDN clock is only 10E-7, which cannot satisfy the

requirement of GSM protocol. BTS adopts internal clock, which accuracy meets the

requirement of GSM protocol.

4) Voice quality

When the transmission bit error is less than 10E-6, the Voice quality is not affected.

Usually, the transmission bit error of satellite circuit is less than 10E-8.

As the link lease is very expensive and the quality is particularly sensitive to

environments, the solution of Abis interface transmission by using satellite transfer

should be positioned for the special areas where the ordinary transmission means is

dissatisfactory and for the emergency communication. When the satellite transfer is

used for networking, the star networking mode is usually adopted. The typical satellite

transfer networking diagram is shown in Figure 2-33.

SDH/PDH/HDSL/Microware

/E1

E1

E1

MSC Earth Station

BTS

BTS

BTS

BTS

Satelite

Earth Receiving

Station

BSC

Earth Receiving

Station

Figure 2-33 Typical satellite transfer networking diagram

Satellite communication is composed of satellite and ground station.

Generally, the satellite communication adopts synchronous satellite, i.e. the satellite

orbit plane is on the equator plane, the satellite is 35786. 6km from the earth surface,

the flying direction is the same as the earth rotation, and the duration of satellite rotation





2-97

cycle is the same as that of the earth. The satellite consists of control system,

communication system (antenna and trunk equipment), telemetry system, power

supply system and temperature control system.

The ground station consists of antenna system, transmitter, receiver, channel terminal

equipment (modem), communication control system and power supply system.

The ground station of ordinary satellite communication is a kind of large-sized

international or European standard communication station. It has such features as high

transmission rate, antenna of large caliber, and expensive cost of equipment. The

subscriber data are connected to the ground station through the ground communication

network to complete communication.

The subscribers in VSAT system form a dedicated network to communicate through

satellite respectively. This mode is featured by its low cost of equipment, antenna of small caliber, and flexible application.

II. Parameter

1) “Transfer Mode ” in [Site\Site Description Table]

2) “Immediate Assignment opt ” in [Cell\Cell Call Control Table]

3) “MS MAX retrans ” in [Cell\System Information Table]

4) “Tx-integer ” in [Cell\System Information Table]

5) “CCCH_CONF ” in [Cell\System Information Table]

2.2.6 Diversity Receiving

I. Technical description

In radio waves propagation, fading (including slow fading and fast fading) may impact

on the communication quality and may even interrupt the communication.

In this technique, the system receives two or more input signals, which carry identical

information but irrelevant random fading features. To minimize these impacts and

enhance the transmission quality, diversity technique is used. It is an effective

approach to overcome fading, encompasses frequency diversity, time diversity,

polarization diversity and space diversity.

1) Space diversity

Space diversity is implemented by providing two sets of stand-alone receiving

equipment concurrently, including antenna, tower amplifier (optional), feeder, DMUX

and receiver. The receiver is made up of two completely independent paths. The input

signals of the two channels come from the master and diversity antennas. The two

signals of space diversity receiving have different propagation environments and

different kinds of fading so they have the feature of coherence or little coherence. It

lowers the impact of propagation factor to adopt diversity combining technology and





2-98

make it output powerful useful signals. In the mobile communication, the wider spacing

interval, the more different multipath propagation, and the less relativity. The interval

between antennas can be either vertical or horizontal. The vertical interval has a poor

performance of diversity, so it is rarely used. In the same BTS or cell, if two sets of antennas with an interval of dozens of wavelength are used to receive the same signal,

the most powerful signals or combined signals with minimum fading can be selected

through diversity combining technology. The diversity gain can be used to indicate the

improvement of space diversity, which value is related with adopted combing

technology. However, the improvement depends on the ratio between the effective

height of diversity antenna (he) and level interval (d), and the incoming wave angle α.

When the frontal signal (i. e. α=0o) is received, the signal coherence coefficient on two

sets of antennas is the smallest one and the gain is the greatest one; when the lateral

signal (α=90o) is received, the coherence coefficient is the greatest one and the gain is

the smallest one.

Space diversity is the most effective and most common mode in the mobile

communication.

2) Time diversity

Time diversity can be used to send the same message through a certain delay, or send

a part of message at different times within the allowed range of delay. Interleaving

technology is used to realize time diversity.

3) Frequency diversity

Frequency diversity is realized through frequency hopping.

4) Polarization diversity

It can get a better diversity gain to set two sets of antenna to form a certain angle.

Moreover, the two sets of antenna can be integrated as one set of antenna. Therefore,

for a sector, only one set of Tx antenna and one set of Rx antenna are needed. If the

duplexer is used, only one set of antenna integrated by Tx and Rx antennas is needed.

Huawei BTS uses dual polarization antenna to realize polarization diversity. This can

realize the combination of antenna, tower top amplifier (optional), feeder, and divider.

When the complicated radio transmission conditions result in deterioration in a path of

the received signals, another path of received signals may vary in signal quality as they

are from an irrelevant transmission path. The BTS receives two paths of signals: main

and diversity signals, demodulates and combines them. This gives 3~5dB diversity

gain.

It has been proven that for the space diversity, a better diversity can be achieved when

the distance between 2 sets of antenna is greater than 10 wavelengths. For the

polarization diversity, it has the advantage of convenient antenna extension and saving

hoist space and is increasedly used.





2-99

II. Parameter

The system needs no extra data configuration to realize the diversity receiving.

2.2.7 Aggressive Frequency Reuse Pattern

I. Aggressive Frequency Reuse

With the development of network, the subscribers increase gradually, the contradict

between short frequency resource and great demand is particularly obvious. In order to

increase capacity, the technology of aggressive frequency reuse should be used to

improve the frequency utilization. According to the actual network circumstance and

requirements, the system can adopt hierarchical aggressive frequency reuse and 1x3

multiplexing technology. The comparison of adopting different aggressive frequency

reuse is as shown in Table 2-13.

Table 2-13 The maximum configuration under different bandwidths

Frequency band 4x3 multiplexing Hierarchical multiplexing 1x3 multiplexing

6MHz S(2/2/2) S(3/3/3) S(4/4/4)

10MHz S(4/4/4) S(6/6/6) S(8/8/8)

Note:

S(4/4/4) indicates three synchronous cells with each carrier number being 4.

In 4x3 multiplexing, 4 indicate four sites, 3 indicates three cells, and totally there are twelve cells as

frequency cluster. Different cells in the same cluster have different frequencies; while cells of other

clusters reuse one certain group of frequency in these twelve frequency clusters.

II. Advanced aggressive frequency reuse technology

1) Hierarchical aggressive frequency reuse

Hierarchical aggressive frequency reuse supports that there can be several different

frequencies multiplexing modes working simultaneously in the same GSM network. For

example BCCH adopts 4x3 multiplexing mode and TCH adopts 3x3 and 2x3 modes.

The nature of hierarchical aggressive frequency reuse is a method of frequency

planning. It has no special requirements of software and hardware for equipment.

Hierarchical aggressive frequency reuse divides all available frequencies into several

groups and each group serves as a carrier layer. The principle of hierarchical

aggressive frequency reuse is as shown in Figure 2-34.





2-100

(1,2,3,4,...36,37)

1 2 3 4 5 6 7 8 9 10 11 12

BCCH TCH1 TCH2BCCH TCH1 TCH2 TCH3 MICRO

Figure 2-34 Principle of hierarchical aggressive frequency reuse

After the hierarchical aggressive frequency reuse is used, frequency hopping, DTX and

dynamic power control should be started to improve C/I thus to satisfy the requirement

of C/I>12dB. The frequency hopping can get the frequency diversity gain and

interference diversity gain.

For example: the maximum configuration S (4/4/4) packet mode can be divided into:

BCCH, TCH1, TCH2 and TCH3.

There are two modes of carrier packet:

Continuous packet: The ARFCNs of frequencies assigned in the same layer are

continuous.

Interval packet: The ARFCNs of frequencies assigned in the same layer have

intervals.

The following examples illustrate these two packets. Provided that frequency range is

512~561, totally 50 frequencies. 12 frequencies are assigned for BCCH, 38 for TCH.

a) Continuous packet mode

BCCH (12): 512~523;

TCH (38): 524~561.

b) Interval packet mode

BCCH (12): 512, 514, 516, 518, 520, 522, 524, 526, 528, 530, 532, 534;

TCH: 513, 515, 517, 519, 521, 523, 525, 527, 529, 531, 533, 535~561.

Both these two packet modes have their advantages and disadvantage. The

comparison is made as follows.

In the case of continuous packet, the interference between BCCH carrier layer and

TCH carrier layer is little. However, both same frequency and adjacent frequency

interference should be considered as a restriction for the planning of BCCH layer.

Meanwhile, BCCH layer and TCH layer are quite independent and there is only one





2-101

frequency between BCCH and TCH layers, therefore, BCCH layer can be easily

modified without interference to TCH layer.

The employment of interval packet mode can guarantees that there is no adjacent

frequency interference between BCCHs. Moreover, the planning of BCCH carrier layer

is relatively easy since the same frequency interference is considered as a main

restriction. However, the interference between BCCH and TCH layers is strong.

Therefore, the planning of TCH layer after the planning of BCCH layer becomes difficult.

Under the condition of the same number of frequencies, the continuous packet mode of

BCCH carrier layer is more difficult than the interval packet mode, for more

consideration should be given to the restriction of adjacent frequency interference. (In

the system with frequency hopping adopted, the less consideration can be given to the

restriction of adjacent frequency interference.

The principle for different carrier layer multiplexing ratios: Assign frequency layer by

layer, try to apply different multiplexing ratios for different layers, and realize aggressive

frequency reuse layer by layer.

General principle: BCCH>TCH1>TCH2>TCH3

When multiple frequency multiplexing is adopted, C/I value will be decreased due to the

aggressive frequency reuse being adopted for each TCH layer. Then the requirement

that the same frequency interference C/I is greater or equal to 12dB in GSM system is

not guaranteed. Moreover, the different frequencies have different interference

situations. The less frequencies in the layer, the more serious interference. If frequencyhopping and other measures are not adopted, the above-mentioned interference

between frequencies will take place thus the communication quality is not guaranteed.

Therefore, the system must adopt measures such as frequency hopping, discontinuous

transmission, and dynamic power control to minimize these kinds of interference.

Frequency hopping can get the frequency diversity gain and interference diversity gain

so as to avoid Rayleigh fading and same frequency interference.

It should be noted that the purpose for different carrier layers using different

multiplexing ratios is to avoid interference at most. This is shown in the flowing aspects.

Under the circumstance of non-uniform network sites, it is not the case for every cell to

use the TRX of last layer or the most last layers. So the TRX of last layer or the most

last layers can realize a higher aggressive frequency reuse ratio (even without the

employment of frequency hopping).

Since the system tries to s are tried to use different multiplexing modes for each carrier

layer, frequencies of any two cells in network are not completely the same, i. e. there is

no the real same frequency cell. After multiple frequency multiplexing is realized,

though interference is increased, the TRX also increased. This makes more

frequencies to participate in frequency hopping and the gain is increased. If the

frequency with weak interference and frequency with strong interference coexist in the





2-102

same cell, they will be mixed after frequency hopping is adopted. The system can still

use the interfered frequencies according to the feature of decoder.

So for each burst, C/I is changeable. But for a specified connection, its quality depends

on C/I equalizing value and the equalizing value is not fluctuated.

III. 1x3 frequency multiplexing technology

1x3 frequency multiplexing technology is a kind of aggressive frequency reuse. The

following is a simple example to illustrate the principle of 1x3 frequency multiplexing.

Provided that the maximum configuration site is S (8/8/8) in a certain area, the available

frequency band is 14. 4MHz, 9 frequencies are reserved for micro-cellar, 12. 6MHz is

left, and there are totally 63 frequencies. Among these 63 frequencies, 15 are assigned

for BCCH carrier (the assignment on the frequency is continuous), and 48 TCH

frequencies are left. Then frequencies are divided into 3 groups (combiner hopping

mode is adopted):

Group 1: 29, 32, 35, 38, 41, 44, 47, 50, 53, 56, 59, 62, 65, 68, 71, 74;

Group 2: 30, 33, 36, 39, 42, 45, 48, 51, 54, 57, 60, 63, 66, 69, 72, 75;

Group 3: 31, 34, 37, 40, 43, 46, 49, 52, 55, 58, 61, 64, 67, 70, 73, 76.

1x3 frequency multiplexing has the advantage of high frequency efficiency, easy

planning method, and easy assignment of frequency. Meanwhile, HSN and MAIO

should be carefully planned and the BTS should support radio frequency hopping. Inlarge cities, there are many BTSs and the site is complicated. The employment of 1x3

frequency planning method can greatly reduce workload and good performance can be

achieved in the case of small multiplexing ratio. 1x3 multiplexing uses the principle that

the number of FH frequencies is greater than the number of carrier frequencies in the

cell to avoid interference and to reduce same frequency collision probability. For a

specified connection, its quality depends on C/I equalizing value. It has been proven

that whether the C/I is good or not depends on same frequency collision probability

after radio frequency hopping. And the collision probability is only related with the

frequency utilization. 1x3 frequency multiplexing mode is as shown in Figure 2-35.





2-103

Figure 2-35 1x3 frequency multiplexing mode

The frequency planning of 1x3 frequency multiplexing is easy and practical, as well as

some disadvantages. For example: when sites are distributed irregularly and the

landforms are greatly different, the collision probability will be greatly increase

Moreover, in the network in which 1x3 planning is implemented, there is also

requirement for network load. When TCH multiplexing ratio is over 40% and the load is

over 80%, the network quality will be decreased rapidly. If the TCH multiplexing ratio is

higher, for example, over 50% and the load is over 60% or 70%, the network quality will

also be decreased rapidly.

IV. Applied conditions for aggressive frequency reuse

To adopt aggressive frequency reuse to improve the frequency utilization and the

network capacity, a series of anti-interference measures should be taken to reduce the

same frequency and adjacent frequency interference caused by aggressive frequency

reuse.

According to the specifications, carrier interference ratio index (engineering value) is:

Same frequency carrier interference ratio: C/I is greater than or equal to 12dB;

Adjacent frequency carrier interference ratio: C/I is greater than or equal to-6dB;

Carrier interference ratio when carrier has an offset of 400 kHz: C/I is greater than or equal to -38dB.

Currently, the following measures are taken to improve the network anti-interference

capability so as to satisfy the carrier interference ratio index: Frequency hopping, DTX

and power control. The following introduces the effect on improvement of network

same frequency C/I and adjacent frequency C/I by frequency hopping.

Frequency hopping has two functions: frequency diversity and interference diversity.

The frequency diversity gain of frequency hopping depends on propagation

environment, MS speed, frequency number of frequency hopping sequence number,





2-104

and the inter-frequency relativity. It is no greater than 6dB. When MS has a high,

frequency hopping has no function of frequency diversity. Generally, the

electromagnetic wave of mobile communication consists of direct wave component and

scattered wave component. When direct wave is in a dominant position, the frequencydiversity of frequency hopping is not obvious. Its gain is about 0~3dB. On the contrary,

when scattered wave in a dominant position, the gain is obvious, which is about 3~6dB.

For a typical environment in which propagation environment, MS speed, and interval

between frequencies are satisfied to achieve the maximum FP frequency gain, the

maximum gain for three-frequency hopping reaches 3. 3dB, 6dB for four-frequency

hopping, no greater than 5.5dB for 9-frequency hopping. The maximum gain of

frequency diversity is no greater than 6dB.

The interference diversity capability of frequency hopping is related to interference

distribution, frequency number of frequency hopping sequence number, and theinter-frequency relativity. Generally, for the narrow band interference, interference

diversity functions apparently; for the broadband interference, it does not function

apparently. It has been proven that when interference is distributed as narrow band and

the number of FH frequencies is 3, 5, 7, the interference diversity gain for interfered

frequency is 3.2 dB, 4.6 dB , 5.5 dB respectively. The function of interference diversity

is shown on the equalization of interference. Therefore, the interference diversity gain

for a single frequency is 0 by default and is sent in the system information

2.2.8 Multiband Network

I. Overview

The multiband network is a network combined GSM900 and GSM1800 In the

multiband network, GSM multiband MS can communicate in either GSM900 frequency

band or GSM1800 frequency band. Each cell in a multiband network has frequencies

from only one frequency band. The multiband network allows cell reselection,

distribution and handover between GSM900 cell and GSM1800 cell. The application of

multiband network is as shown in Figure 2-36.





2-105

GSM1800 Cell

BSC

MSC

BSC BSC

MSC

BSC

GSM900 GSM1800

GSM900 Cell

Figure 2-36 GSM900/GSM1800 multiband network

The multiband network can be used to utilize the abundant frequency resources in

GSM1800 frequency band, to absorb network traffic, and to satisfy the increasing

demand of network capacity.


To guarantee the stable operation of multiband network, it is of utmost importance to

correctly configure the parameters related with the multiband network operation at the

stage of network commissioning. Given below is a description of the technical

principles governing the multiband network.

1) MS Classmark

In the GSM system, MS Classmark represents the MS services, supported bands,

power, and encryption capability. The Classmark of the mobile station falls into three

categories: Classmark1, Classmark2 and Classmark3. The network can interrogate the

Classmark of MS and realize its capabilities. In addition, the network can request the

mobile station to report its Classmark3 immediately after creating a link by setting the

parameter “Early Classmark Sending Control”. Since the important messages in

Classmark3 are created specially for multiband applications, it is required that in the

multiband network the equipment should support the processing of MS Classmark.

Huawei BSS supports ECSC, processing of MS Classmark3, etc.

2) BA list

In the GSM system, the BA (BCCH Allocation) list is a set of all the carrier channel

numbers of adjacent cells of each cell.

The network carries out compatibility handling for various types of MSs through system

information control. It also guides the MSs to access and handover correctly so that

good services of the radio network can be guaranteed.





2-106

BA defines the absolute channel numbers used by carrier of all adjacent frequency cell

BCCHs, which is used for cell selection and handover. It is the system that informs MS

the BA list through system information. There are two types of BA list:

The BA1 mainly contains the list of adjacent cells searched by the MSs in idle

mode. It is transmitted periodically in the system information type 2, 2bis or 2ter,

and used for cell re-selection in the idle mode.

The BA2 mainly contains the list of adjacent cells searched by the MSs in active

mode. It is transmitted in the system information 5, 5bis or 5ter, and used for

handover in active mode.

When an MS is in active mode, It extracts the parameters of adjacent cells from the

associated channel system information type 5, 5bis or 5ter on the SACCH, instead of

from the system information type 2, 2bis and 2ter. In accordance with the actual

network status, the BA list in the system information type 5, 5bis and 5ter can be either identical to or varied with that in the system information type 2, 2bis and 2ter.

The BA list shall be set in accordance with the network design requirements and the

actual status of adjacent cells. Otherwise, there might be inappropriateness in

handover or cell re-selection, or even handover failure. In this case, it may impact the

services provided by the network.

The number of adjacent cells on each BA list shall not exceed 32.

3) Support of system Information for multiband network

The network carries out compatibility handling of MSs of various classes throughsystem information (type 2 / 2bis / 2ter and 5 / 5bis / 5ter). The radio network controls

the MSs to access and handover correctly and guarantees good services.

Huawei GSM system carries out thorough compatibility processing of Phase 1 and

Phase 2 900 MSs, Phase 1 and Phase 2 1800 MSs and multiband MSs, and supports

system information type 2 / 2bis / 2ter and 5 / 5bis / 5ter.

The BA1 list is sent in system information type 2 for re-selection. The BA2 list is sent in

the system information type 5 for handover. In GSM900 system, the frequency

channels are numbered from 1 to 124. Coding can be done on one list without 2bis /

2ter / 5bis / 5ter when the bitmap format is used. However, this should be adjusted after

the multiband system is employed.

For the GSM900 cells, the GSM1800 frequency channels on its adjacent cell list are for

multiband MSs. They are transmitted via the system information type 2ter / 5ter. Only a

multiband MS supports the system information type 2ter / 5ter. Whereas the frequency

channels of its GSM900 adjacent cells are placed in the system information type 2 / 5

and can be coded in the bitmap format. The Phase 1 MS recognizes the bitmap format

only. This ensures compatibility with Phase 1 GSM900 MSs.





2-107

For the GSM1800 cells, they are handled in a similar way. The 900M frequency

channels on the list of its adjacent cells are for multiband MSs, transmitted through the

system information type 2ter / 5ter. Whereas the frequency channels of its GSM1800

adjacent cells are placed in the system information type 2/5. As they cannot be codedon one list, the BA list needs to be split into two parts, transmitted respectively in the

system information type 2 (or 5) and 2bis (or 5bis). The system information type 2bis

(5bis) is for single-band M1800 MSs and multiband MSs only.

For the multiband network, it is required that the equipment should support the system

information type 2ter/5ter.

4) ECSC (Early Classmark Sending Control)

ECSC indicates whether MS is required to report the MS Classmark3 voluntarily and

early. For further details refer to the protocol 0408. On receipt of the Classmark change

message, MS will send the additional Classmark message to the network as early as

possible. Classmark3 information includes the power messages of various frequency

bands of multi-frequency MS. In the handover between different frequency bands,

power level should be correctly described. It is essential to know the Classmark3

message when making a paging call or sending the BA2 list in different bands.

The sampling range of the ECSC is as follows:

ECSC=1, transmission required.

ECSC=0, transmission not required.

This function comes into beings for the multiband-networking situation. And the

information in Classmark3 is for multiband application.

In case of single-band networking, it is advisable to set this parameter to 0. In case of

multiband networking, the recommended value of ECSC is 1 so that signaling flow can

be reduced.

The parameter ECSC is transmitted in system information type 3.

5) MBR (Multi-Band Report)

MBR serves to help the network to notify the MS that the 6 adjacent cells reported mustcover multiple bands.

In the single-band GSM system, when the MS reports the adjacent cell measurement

results to the network, it need only report the 6 adjacent cells with strongest signals in a

band. When there is a multi-band network, the operator will usually expect the MS to

have the priority to enter a band in time of handover depending on the actual status of

the network. Therefore, the MS is expected to report the measurement results based

not only on the level of the signals but also on the band of the signals. The system

parameter “Multi-band Report”, therefore, serves to notify the mobile station to report

the multi-band adjacent messages.





2-108

In the multiband network, the following situation often occurs because the propagation

loss in the 1800 MHz band is larger than that of the 900MHz band: among the 6

adjacent cells with strongest signals as reported by MSs, none of them is a GSM 1800

cell. This will affect the absorption of traffic by the GSM1800 network. In this case, thenetwork can request the multiband MSs to send the MR about the adjacent 1800MHz

cells by setting the MBR value. By setting different values for MBR, the MSs can report

the messages of the adjacent cells of different bands as required when submitting the

MRs of 6 best adjacent cells.

MBR is represented in decimal digits, with the ranges from 0 to 3. Its implication is

shown in Table 2-14.

Table 2-14 MBR implication

MBR Implication

0MS shall report the measurement results of 6 adjacent cells with strongest signals known andallowed by NCC depending on the signal level of the cells, regardless of which band the cellsare in.

1

MS shall report the measurement results of an adjacent cell in each band with strongest signals,which are known and allowed by NCC on the adjacent cell list. Then it shall report the adjacentcells in the band used by the current service area in the remaining space of the report. If there isstill space left, it shall report the status of the other adjacent cells, regardless of which band theyare in.

2

The MS shall report the measurement results of two adjacent cells in each band with strongestsignals known and allowed by NCC on the adjacent cell list. Then it shall report the adjacent

cells in the band used by the current service area in the remaining space of the report. If there isstill space left, it shall report the status of the other adjacent cells, regardless of which band theyare in.

3

The mobile station shall report the measurement results of three adjacent cells in each bandwith strongest signals known and allowed by NCC on the adjacent cell list. Then it shall reportthe adjacent cells in the band used by the current service area in the remaining space of thereport. If there is still space left, it shall report the status of the other adjacent cells, regardless of which band they are in.

6) PI (Cell Reselection Parameter Index)

PI is used to notify MS whether to adopt C2 as cell reselection parameter and to

calculate whether the parameter of C2 exist.

Value range of PI: Y or N.

Y indicates that MS should extract parameters from broadcasting of system information

in cell to work out C2 value and use the value to serve as the standard of cell

reselection; N indicates that MS should use C1 to serve as cell reselection standard (i.

e. C2 = C1). Generally, PI is set as Y in multiband network.





2-109

III. Traffic guide strategy in multiband network

In the multiband networking, one of the most important purposes is to try to let

GSM1800 network absorb or share traffic so as to satisfy the increasing requirement of

network capacity and quality. The following principles should be followed.

In the early stage of multiband network construction, try to let GSM1800 cells

absorb multiband subscribers.

Realize the continuous coverage of GSM1800 network in hot spot areas.

When the number of multiband subscribers reaches a certain level, use different

bands to share traffic thus to reduce handover and provide better service.

The carrier can realize different traffic control strategy through real-time adjustment of

related parameters.

Different traffic control methods are used for different MS states. GSM1800 cell can

have higher priority or better adjacent cell measurement comparison value through the

configuration of system parameters. So when the subscriber turns on the mobile to

select cell in idle mode or reselects cell in standby state, GSM1800 cell can be more

likely to be the serving cell for multiband subscribers. In this way, the subscriber is more

likely to wait at GSM1800 before a call connection; during the connection of MS call,

the traffic distribution can be adjusted by directed retry. In connected state, try to

connect as much as possible traffic to high level GSM1800 cells in lower layers through

cell hierarchy and specifying different hierarchical cell structures (HCS); the multiband

traffic handover can be used to make traffic load more rational.

The following describes in detail the cell selection, cell reselection, directed retry, cell

hierarchy and specifying HCS, and multiband handover.

1) Cell Selection and Cell Reselection

In idle mode, the system guides the traffic absorption by controlling the process of MS

cell selection and cell reselection.

When MS turns on, it first needs to select cell so that to confirm its serving cell. Principle

of cell selection: cells allowing to be accessed and cells with high priority are first

selected; for the cells with the same priorities, the cell with maximum C1 value is first

selected. The C1 value of selected cell should be greater than zero. C1 value is

calculated as follows:

( )( )0, _ _ _ _ _ 1 P CCH MAX TxPWRMS MAX MIN Access RxLEV RxLEV C −−−=

RxLEV Access MIN range: 0~63, 0 is corresponding with -110dBm, 63 is

corresponding with -47dBm.

MS TxPWR MAX CCH value range:

GSM900: 0~19 available, 0 is corresponding to 43dBm, 1 is corresponding to

41dBm. The value of higher level is 2dB greater than that of lower level.





2-110

GSM1800: 0~15 available, 0 is corresponding to 30dBm, 1 is corresponding to

28dBm. Step is 2dB.

In the multiband network, owing to the strong fading of signals in GSM1800 frequency

band, signals in GSM900 frequency band is stronger. In order to enable MS can be

accessed to GSM1800 system, the cell selection priority can be controlled by setting

value of cell bar qualify (CBQ) and cell bar access (CBA). The signals in GSM1800 cell

are generally weaker than that in GSM900 cell. To enable the multiband MS to select

GSM1800 cell preferentially, GSM1800 cell can be set as Normal and GSM900 cell as

Low.

Table 2-15 Cell selection/reselection hierarchy

Case CBQ CBA Cell selection Cell reselection

1 0 0 Normal Normal

2 0 1 Barred Barred

3 1 0 Low Normal

4 1 1 Low Normal

When selecting cell, GSM900 cell is set as CBQ=1, CBA=0 and GSM1800 is set as

CBQ=0, CBA=0. This enables GSM1800 cell to have a higher priority.

After MS completes cell selection, it should reselect cell in standby state in order toselect a better serving cell. The parameter that decides cell reselection is C2. MS

reselection principle is to select the cell with maximum C2 value as the serving cell. C2

depends on the following factors:

C2=C1+CRO-TO×H(PT-T) (PT<31)

C2=C1-CRO (PT=31)

Where, the value Cell Reselection Offset (CRO) decides the difficulty of cell reselection

and Temporary Offset (TO) functions within penalty time (PT).

CRO value can be 0, 1, ↑ 63 with grade as unit, which are corresponding to 0=0dB;

1=2dB; 63=126dB respectively.

TO value can be 0, 1, and 7, which are corresponding to 0=0dB; 1=10dB; 6=60dB;

7=infinite respectively.

PT value can be 0, 1, and. 31, which are corresponding to 0=20s, 1=40s, and 30=620s

respectively.

H ( ) = 0 if PT-T<0

H ( ) = 1 if PT-T>0





2-111

C1 indicates the quality of radio channel. The greater C1 value, the better quality of

channel. C2 is corrected manually. C2 value of each cell can be adjusted through CRO

value. So the C2 value can be calculated according to CRO, TO, and PT so as to

confirm the cell reselected for MS. That is to say, C2 value of GSM1800 cell can begreater than that of GSM900 cell by setting parameters that can affect C2 value, such

as CRO. Therefore, though signals in GSM1800 cell are weaker that of GSM900 cell,

GSM1800 cell can still be reselected for MS by setting parameters.

Parameters of cell selection and reselection can be flexibly used to control MS to select

GSM1800 network as required in network planning; under the precondition that

network quality is guaranteed, these parameters can be used to make MS establish

calls in GSM1800 network so as to share the load of GSM900 network.

2) Directed retry

Provided that the process to initiate a call by an MS has completed switching,

connection, control of some signaling and it is time to for SDCCH to assign TCH so as

to connect the speech channel of both parties. However, it is found that the TCH of this

cell is full. In this case, directed retry can be used to assign TCH of adjacent cells for

MS from SDCCH thus to guarantee the successful connection. At the same time, the

traffic is shared.

3) Layers and levels of network

Under the connected state, traffic between frequency bands can be distributed

rationally through abundant Huawei multiband handover This is the core of multiband

traffic guide and control strategy.

Huawei hierarchy handover algorithm divides a cell into 4 layers each layer with 16

levels. This meets the need of complicated networking circumstances. The design

concept of this hierarchy has fully considered the collaboration with the current network

equipment and the requirement of future network development. The cell layers and

levels are as shown in Figure 2-37.





2-112

GSM 900

GSM900Cell

Layer 4

Layer 3

Layer 2

Umbrella

Cell

GSM 900 GSM 900 GSM 900

GSM900GSM900

GSM1800GSM1800

GSM900 GSM900

GSM1800 GSM1800

Micro CellLayer 1

GSM 1800 GSM1800 GSM1800GSM 1800

Cell

Figure 2-37 Cell layers and levels

The GSM system covering the same area is divided into 4 layers. The high layer is the

fourth layer, i. e. umbrella-like cellular, which is generally is a GSM900 cell with wide

coverage. It has two functions: covering and quick connection of MS. The middle layer

consists of GSM900 macro cells. These are the main cells of the system and most of

subscribers gather in this layer. The followed layer consists of GSM1800 micro cells

with small coverage. This layer is the main target for capacity expansion so as to solve

the problem of short resource of frequencies. The bottom layer consists of GSM1800

Pico cells, which is to meet the requirements of hot spot and blind spot areas. For the

priority, the cell in lower layer has a higher priority.

Considering the future network development, to make network planning and

optimization more detailed and more flexible, the layer should combine with level

division, that is to say, each layer should be divided into several levels. Each layer of

these four layers is divided into 16 priorities.

For the description of handover, please refer to 2.3.6 .

IV. Features of GSM1800

1) Propagation characteristics of GSM1800

The working frequency of GSM1800 is two times as that of GSM900. According to

COST-231 model and practical experience, the propagation loss of GSM1800 inside

stadia is 6dB greater than that of GSM900 and the propagation loss of GSM1800

outside stadia is 10dB greater than that of GSM900. The propagation loss inside

buildings is 5~17dB higher (it varies from material to material). The fast fading of

GSM1800 is a disadvantage to realize the fine coverage of GSM1800 and the condition

of GSM1800 coverage is directly related with the performance of network. Moreover,

electromagnetic diffraction of GSM1800 is poorer that than of GSM900.

2) GSM1800 coverage requirements





2-113

a) Outdoor coverage

The outdoor coverage can be easily realized if the distance between sites is not too far.

If necessary, besides the installation of GSM1800 equipment on the site of original

GSM900 site, the new sites should be installed in proper places.

b) Indoor coverage

In order to guarantee the fine indoor coverage of GSM1800, the distance between

BTSs in the city should not exceed 1000m. In the city with buildings in reinforced

concrete structure, which penetration loss is very great, so it is recommended that the

distance between BTSs should be 500~800m.

3) GSM1800 coverage mode

There are three coverage modes for GSM1800 network in multiband network: Fine

continuous coverage, continuous coverage of hot spot areas, scattered coverage of hot

spot areas.

a) Fine continuous coverage

This coverage mode has the following advantages: GSM1800 is easy to absorb traffic

and has less handovers and high quality of operation; the frequency planning and

network optimization is easy to be realized and the traffic distribution is easy to be

controlled; after sites are constructed, if capacity expansion is needed, it is only needs

to configure carrier instead of constructing new sites; and it is convenient to be

constructed and maintained. The disadvantage is that the investment is large and it ishard to select sites in one time.

b) Continuous coverage of hot spot areas

This coverage mode has the following disadvantages: the traffic absorption of

GSM1800 is limited and there are frequent multiband handovers; strict requirement for

locating traffic hot spot; it is hard to plan frequencies and optimize network due to the

irregular distribution of GSM1800 BTSs. The construction and maintenance is

complicated. The advantage is that the site in highly intense areas can be gradual

constructed so as to save the investment.

c) Scattered coverage of hot spot areas

This coverage mode has the following disadvantages: the traffic absorption of

GSM1800 is low and there are frequent multiband handovers; strict requirement for

locating traffic hot spot; it is hard to plan frequencies and optimize network due to the

irregular distribution of GSM1800 BTSs. The construction and maintenance is

complicated. The advantage is that the initial investment is small.





2-114

V. Multiband networking modes

There are three modes for multiband networking: standalone MSC networking, shared

MSC/standalone BSC networking, and shared BSC. In general, the former two modes

are called standalone networking and shared BSC network is also known as mixed

networking.

1) Standalone MSC networking

Standalone MSC networking refers to that GSM900 and GSM1800 use different MSCs

for networking respectively, as shown in Figure 2-38.

M S

M S

B T S

B T S

B S C

E IR

H L R / A U C

O M C

S M C

G S M 9 0 0 G S M 1 8 0 0

B T S

B T S

B S C M S C / V L R

M S C / V L R

Figure 2-38 Standalone MSC networking mode

The standalone MSC networking has the following features:

No impact on original network

Clear network planning, clear network data configuration, and easy to construct.

Satisfy the requirement of long-term capacity expansion.

Convenient to manage the whole network and develop new services.

The initial investment of network is relative large but the investment for each

subscriber is the smallest.

Introduce competition so as to lower equipment investment and improve quality of

service.

Besides the above features, the standalone MSC networking increases the inter-office

handovers and position updates, thus the load of signaling link is increased. In addition,

the standalone MSC networking has the problem of collaboration of equipment of

different providers. In a long-term view, it is better than mixed networking.

2) Shared MSC / standalone BSC networking

Shared MSC / standalone BSC network refers to GSM900 and GSM1800 network

adopts the same MSC and different BSCs for networking, as shown in Figure 2-39.





2-115

MS

MS

BTS

BTS

BSC

EIR

HLR/AUC

OMC

SMC

GSM900 GSM1800

BTS

BTS

BSC

MSC/VLR

Figure 2-39 Shared MSC/standalone BSC networking

Shared MSC/standalone BSC networking has the following features:

It has impact on the original network.

It needs to plan NSS again and it is hard to be constructed.

It is hard to expand its capacity. As network develops, construction and maintenance

might become difficult.

The initial investment of network is relative small and the investment for each

subscriber is small.

Introduce competition so as to lower equipment investment and improve quality of

service.

BSC has backup function so the network security is good.

3) Shared BSC networking

Shared BSC networking refers to that BTSs of GSM900 and GSM1800 access the

same BSC or multiband mixed BTS accesses BSC, as shown in Figure 2-40.

MS

MS

BTS

EIR

HLR/AUC

OMC

SMC

GSM900 GSM1800

BTS BSC

MSC/VLR

BTS

BTS

BTS

BTS

BSC

BTS GSM1800/GSM900

Figure 2-40 Shared BSC networking mode





2-116

Shared BSC networking has the following features:

It might impact on original network greatly, especially when BSC has a small

capacity.

It needs to plan NSS and BSS again so the construction is difficult.

It is hard to expand its capacity. As network develops, construction and

maintenance might become difficult.

Development of new services is restricted.

It cannot introduce competition thus it is hard to lower the cost and to improve

service.

The initial investment of network is the smallest and the investment for each

subscriber is the largest.

2.2.9 Carrier Mutual-assistance

I. Overview

In case of BCCH TRX failure or baseband frequency hopping TRX failure, the cell can

handle it automatically through the TRX aiding function. Thus, the cell services can not

be affected before the failed TRX is replaced.


TRX aiding contains BCCH TRX aiding and baseband frequency hopping TRX aiding.

For the non-baseband frequency hopping cell, only BCCH TRX aiding will occur. For

the baseband frequency hopping cell, both BCCH TRX aiding and baseband frequency

hopping TRX aiding may occur.

1) BCCH TRX aiding

In the idle state, MS needs to know some information about the infrastructure of the

network. BSC sends the generic broadcast message to BTS, and BTS broadcasts it on

BCCH. BCCH is a low-capacity channel and can send a message of 23 bytes every

0.235s. The broadcast information includes cell selection information, adjacent cell

information, access control information, private channel control information, cell

identification code, location, system parameters of packet service, etc.

When BCCH TRX of a cell is failed, all services of this cell will be interrupted. In order to

ensure the cell services not to be affected, in case of BCCH TRX failure, another

available TRX of the cell can substitute for the TRX that BCCH is originally on. Thus,

the cell can continue to provide the services. After the fault of TRX that BCCH is

originally on is removed, BCCH can be recovered (or, changed back) onto this TRX.

This is the function of BCCH TRX aiding.

2) Baseband frequency hopping TRX aiding





2-117

In the baseband frequency hopping cell, if a TRX participating in frequency hopping is

failed, the conversations on this frequency hopping channel will lose some voice

frames. Correspondingly, the communication quality will be decreased. In order to

ensure the communication quality, in case of baseband frequency hopping TRX failure,BSC will start the TRX aiding function. It will automatically change the cell to the

non-frequency-hopping mode. Thus, the failure of a TRX will not affect the

communication quality of the entire cell. When the fault is removed, this cell can be

restored to the frequency hopping mode. This is the function of baseband frequency

hopping TRX aiding.

When TRX aiding or TRX aiding recovery occurs, there will be corresponding alarms

reported (all are event alarms):

198: BCCH TRX aiding alarm.

199: BCCH TRX aiding recovery alarm.

200: Baseband frequency hopping TRX aiding alarm.

201: Baseband frequency hopping TRX aiding recovery alarm.





2-118

Note:

If BCCH TRX in a baseband frequency hopping cell is failed, except for BCCH TRX aiding, baseband

frequency hopping TRX aiding will also occur. That is, the cell is changed to the non-frequncy-hoppingmode. In addition, only when the faults of all TRXs participating in frequency hopping and the original

BCCH TRX are removed, can the baseband frequency hopping TRX aiding recovery occur. That is,

the cell is restored to the frequency hopping mode.

When TRX aiding or TRX aiding recovery occurs, the cell will be initialized again.

In the previous BSC versions of G3BSC32.10100.06.1120A, the TRX aiding function cannot be used

together with the baseband timeslot frequency hopping function. From the version of

G3BSC32.10100.06.1120A, this limitation is canceled. These two functions can be used at the same

time.

If such adjustment as SDCCH dynamic adjustment, full-rate/half-rate dynamic adjustment or PDCH

channel dynamic adjustment occurred in the cell, the dynamic adjustment channel on the TRXs

involved in TRX aiding will be restored to the channel type configured originally. For BCCH TRX aiding,

the involved TRXs contain the current BCCH TRX and TRX to be aided. For baseband frequency

hopping TRX aiding, the involved TRXs are all TRXs in the entire cell.

When inter-E1 BCCH TRX aiding occurs, the traffic channels of the current BCCH TRX can not work

normally if the E1 carrying the original BCCH TRX is broken. From the version of

G3BSC32.10102.06.1120A, BSC will automatically block the traffic channels of the current BCCH

TRX when the E1 carrying the original BCCH TRX is broken. These channels will be unblocked when

that E1 recovers.

III. Parameters

The TRX aiding function only uses a parameter for controlling. The parameter is

configured in [Cell Configuration Data Table], as shown in Table 2-16.

Table 2-16 Description of parameter of Cell Configuration Date Table

Parameter Value range Description

TRX Aiding Not Allowed TRX aiding is not allowed. That is, the TRXaiding function is closed.

Allowed, Recover Forbidden

TRX aiding is allowed. However, after thefault TRX is restored, TRX recovery isforbidden.

Allowed, Recover Immediately

TRX aiding is allowed. After the fault TRX isrestored, it can be recovered immediately.

TRX Aiding FunctionControl

Allowed, Recover WhenCheck Res (default valueof the field).

TRX aiding is allowed. After the fault TRX isrestored, it will not be recoveredimmediately but recovered during resourcecheck at 3:00 am.





2-119

2.2.10 Cell Broadcast

I. Overview

Cell broadcast is a specific service of the GSM system, broadcast information to all

mobile stations in a specific area periodically. The MS supporting this service can

monitor this broadcast information continuously and this information can be displayed

on the MS terminals. The typical examples of cell broadcast are to broadcast traffic

information and weather forecast.

Short Message Service Cell Broadcast (SMSCB) allows short message to be

broadcast to all mobile stations in certain areas. These areas may be one or several

cells, even the entire PLMN area. The short message from Cell Broadcast Centre (CBC)

is sent to BSC, and BSC will manage and dispatch the message, and send the receivedmessage to BTS, which can control the flow of short message broadcast.

The architecture of cell broadcast system is as shown in Figure 2-41.

CBC

CDB

CDB

BSC

BSC

BTS

BTS

BTS

OMC CBC

GMEM

GMEM

...

...

Remote connection

LAN connection

GMEM

GMEM

WAN

LAN

Figure 2-41 Cell broadcast system architecture

II. Cell broadcast functions

1) Receiving and storing of short messages

CDB receives and stores the short message from CBC. There are three kinds of

commands to broadcast short messages sent from the CBC: send a new broadcast

short message, delete an outdated message or a message that meet specified

requirements, and replace an old message with a new one. CDB handles these three

cases respectively and updates the memory of short messages, i. e., adding the new

message to the short message database. On the reception of new command it deletes

the older one, or deletes the older message before adding a new message after the

reception of replacing message. If old message can not be deleted then the new

message will not be added.

2) Dispatching and transmitting of short messages





2-120

Every short message should be broadcast in specific areas, which correspond to one

or more cells. In the meantime, every message has its own transmission requirements,

i. e. the transmit times and frequencies are different. BSC should send each message

to the specific area according to its transmit requirements. When a cell requestsmultiple short message, BSC should calculate the transmitting time sequence

according to specific message dispatch algorithm, and send these messages to BTS in

turn according to the assigned sequence.

3) Responding to the query of cell broadcast centre

While storing and transmitting short messages, CDB will record the completion

message, the number sent by each cell, message load conditions of each cell and the

broadcast channel state of each cell. CBC can keep track of the current system running

state by querying and monitoring the cell broadcast system. It can adjust and optimize

the system to ensure satisfactory running. When fault occurs in cell, CDB will report it toCBC, which will stop sending short messages to this cell. When CBC identifies that a

specific broadcast message is been sent, it will send a command to delete or replace

this message from CDB to reduce its load.

III. Cell broadcast features

1) Supporting MS DRX mode

If MS in idle mode has selected its serving cell, it is ready to monitor the paging

message from this cell. To lower the power consumption of MS, the GSM specification

adopts the discontinuous receiving mechanism (DRX), i. e. each subscriber (IMSI)corresponds to a dedicated paging group and each group corresponds to a paging

sub-channel of the cell. MS recognizes its paging group and the corresponding paging

sub-channel according to the last three digits of the IMSI and PCH allocation on service

cell. MS in idle mode uses its own paging sub-channel to receive the paging message

(or to monitor the receiving level of the BCCH carrier of the non-serving cell). MS

ignores the message from other paging sub-channel or even shuts down the power of

some hardware to lower its power consumption during the broadcasting of other paging

sub-channels. But MS must measure the network messages task periodically.

The BSC supporting DRX mode needs to send scheduling messages to satisfy therequirement of the discontinuous receiving by MS. One dispatch message contains the

information of the short messages to be sent one after another in a cell. The cycle

occupied by short messages in one dispatch message is called dispatch cycle.

Dispatch message contains the descriptions of each short message to be broadcast

according to the transmit sequence, and indicates the message position in dispatch

cycle. Mobile subscriber can read concerned short message in less time by reading

dispatch message, thus minimizing power consumption.

2) Supporting traffic control for BTS





2-121

Transmit sequence of short messages in each cell is dispatched by BSC and

transmitted by BTS. Each TRX maintains one message buffer and sends short

message periodically to MS through specific channel. Thus, there will be an

asynchronous state of sending short messages between BSC and BTS. In this case,BTS will report this asynchronous state to BSC in the form of load indication message.

If a specific TRX receives too many short messages and can not send them in time,

BSC will temporarily stop sending short messages of this TRX. If a TRX barely receives

short messages, BSC will send out some short messages so that the time sequence for

sending short message in this cell will be met. By sending broadcast messages to BTS

to control traffic, CDB can schedule the balance of the broadcast system of the whole

cell. Thus the requirement of sending broadcast messages is satisfied as much as

possible.

2.2.11 Radio Channel Allocation

BSC is responsible for the allocation of circuit channel and PCU for the allocation of

packet data channel.

1) Radio channel allocation requirements

Radio channel allocation is based on the following requirements:

Initial channel allocation: An idle MS enters active mode during MOC, paging

response, or location updating.

Connection allocation: The channels allocated can not meet the requirements. For

example, MS has been allocated SDCCH, but if need to transfer speech or data.

Handover: Due to the subscriber mobility or the change in the interference level, it

is necessary to hand the MS call to another channel.

For the dedicated channel allocation management of the BSC system, VEA (Very Early

Allocation) and EA (Early Allocation) are used in a combined manner as an allocation

strategy.

VEA refers to the allocation of TCH at initial stage. EA refers to the allocation of TCH

after the initial allocation of SDCCH.

During channel allocation in the BSC, the VEA is used for some special calls. For

common calls, the EA technique is used, which can effectively improve channel

utilization efficiency.

2) Radio channel allocation algorithm

BSC channel allocation algorithm selects the channel for allocation by considering

channel interference, configuration, history record, load distribution, MS transmitted

power, etc., and based on the specific call event and environment.

Channel interference directly determines such critical traffic statistic indices as the

quality of communication completion ratio and call drop rate on the channel. It is the





2-122

most important factor to be considered during the channel allocation. The measurable

interference includes uplink interference of the idle channel and the uplink / downlink

interference of the occupied channels.

The rule for the interference-based channel allocation algorithm is to select the channel

of lower interference for allocation. But there are two special cases. The first is that for

a high-level call or user, the MSC may have an interference limit for channel allocation.

The channel with interference higher than this limit can not be allocated. The second is

the specific call environment in which the maximum transmitting power capacity of MS

and path loss are considered. The call with better receiving level can be allocated with

a channel having severer interference. The channel with lower interference is reserved

for the call of poorer receiving level and thus the call completion ratio and

communication quality can be improved.

The channel allocation algorithm based on channel configuration is based on the

following factors: whether the carrier of the channel is that of BCCH, the frequency

reuse distance of the TRX, whether the channel use frequency hopping, and the

number of frequency in the frequency hopping group. Proper frequency allocation

based on channel configuration helps to reduce the interference of network, and

improve the quality of network.

The channel allocation algorithm based on channel history record is characterized by

the memory function. The history record includes the channel seizure success or failure

and call drops, and it needs to verify whether the cause of seizure failure and that of call

drop lies in the radio channel itself. Such history records can provide reliable facts for

the current channel allocation.

The channel allocation algorithm based on load balancing is characterized by even

distribution of the carrier frequencies, Time Slots (TS) and sub-TSs during the channel

allocation. It can reduce adjacent-channel interference and same frequency

interference. On the other hand, it also helps to avoid the risk caused by calls being

concentrated on a few carriers.

There are also special allocation methods for specific call events such as intra-cell

handover and IUO handover. For example, intra-cell handover is mainly caused by thequality problem of the speech channel, which indicates that the carrier where the

original channel is located has suffered interference. If the original channel frequency

hops, then some frequency bands in the frequency-hopping group of the original

channel may have suffered severe interference. In case of intra-cell handover, the

allocation of new channel may select the carrier and frequency hopping group that are

different from those of the original channel.

3) Queuing

Channel allocation algorithm used in the BSC supports queuing. In case of initial

allocation, no queuing takes place because an MS will resend the channel request.





2-123

Queuing is mainly applied to connection allocation and handover, where the MSC

decides if queuing is allowed in the allocation request or handover request. If no

allocable radio channel is available and queuing is allowed, then the M900/M1800 BSC

will queue the allocation requests. Try to allocate traffic channel with the allowed periodof time so as to reduce the subscriber’s wait time.

4) Allocation by priority

The channel allocation algorithm in the M900/M1800 BSC supports different priority

levels, that is, channel allocation can be carried out according to the preset priority

levels. In some cases, the request of higher priority can be forcibly implemented and

can occupy the channel, which is being used by a user of lower priority.

5) Dynamic allocation of SDCCH

a) Purpose

The objective of SDCCH dynamic allocation is to optimize the usage of traffic channels

and signaling channels, reduce the occurrence of congestion on the SDCCH, and

lower the impact of the initial configuration of the SDCCH on the system performance.

The number of SDCCHs required is based on the traffic model, that is, the current traffic

distribution and statistical data about handover. An increase of short message service

will lead to the increase of requirement for the SDCCHs, which makes the prediction of

the SDCCHs requirement very difficult.

There may be the case that the number of users in a cell suddenly increases, and many

users can not access the network just because they fail to request the SDCCH. In this

case, TCHs have to be converted into SDCCHs so as to ensure that most of the users

can access the network and communication can be implemented through directed retry

function, which improves the call success rate.

b) Advantages

It's not necessary to work out the exact number of SDCCH in advance after

implementing SDCCH dynamic allocation. SDCCH dynamic allocation increases the

system capacity and improves the call completion ratio.

The disadvantage of SDCCH dynamic allocation is increased intra-cell handover traffic.

It can be overlooked on account of its advantages.

c) Approach

SDCCHs are allocated with the cell as a unit. The following should be configured in the

data management console: dynamic/static allocation of SDCCH, idle SDCCH threshold,

and maximum number of cell SDCCHs, minimum time for TCH recovery, etc.

If SDCCH allows dynamic allocation and satisfy the following conditions:

When number of SDCCHs is less than or equal to idle SDCCH threshold;





2-124

And number of existing SDCCHs in the cell is less than the maximum number of

SDCCHs;

Number of idle TCHs is greater than 4 or greater than the number of configuration

carriers.

The system will automatically select a TCH and turn it into SDCCH. At the same time,

BSC delivers configuration command to BTS to configure this TCH as SDCCH and

update the channel list of internal BSC.

When there are many idle SDCCHs, the SDCCH channels are dynamically converted

into TCHs.

6) Dynamic allocation of PDCH

a) Introduction

To support GPRS service, two types of channels are introduced, i. e. , static PDCHs

and dynamic PDCHs. Static PDCHs are used for packet service only. Dynamic PDCH

is initialized as a TCH and controlled by BSC. When the static PDCHs are not sufficient,

the PCU will apply for dynamic PDCHs from the BSC. When the PCU is granted with

the control authority, dynamic PDCHs are used for packet service. On the contrary, if

TCHs are insufficient, the BSC can request dynamic PDCHs from the PCU. When the

BSC is in control, the dynamic PDCHs serve as TCHs. According to the protocol,

following channel combinations are provided.

PBCCH+PCCCH+PDTCH+PACCH+PTCCH

PCCCH+PDTCH+PACCH+PTCCH PDTCH+PACCH+PTCCH

PBCCH+PCCCH

Huawei BSS supports the first three combinations.

b) Approach

The dynamic PDCH control is based on cell, the following two items should be

configured: Idle TCH threshold N1; application TCH decision period T (minute). Set a

Count for each cell with initial value being 0. Count value ranges -T~T. Adjust Count

every one minute. If the number of the current channels is greater than N1, then 1 is

added to the counter value, if the number of the current idle TCHs is less than or equal

to N1, then 1 is subtracted from the counter value. If the counter value is less than -T/2

after adjustment, then the BSC requests dynamic PDCHs from the PCU. After the BSC

acquires the control power, dynamic PDCHs serve as TCHs. If any change takes place

to the current type of a dynamic PDCH, it is necessary to issue a configuration

command to the BTS so as to configure this channel as the current type and update the

channel list in the BSC.

c) Note:





2-125

It is difficult to predict the packet traffic of the cell. The introduction of dynamic PDCH

can improve the utilization of the channels.

Channel allocation follows the principle that circuit switching service being preferred to

packet switching service. The PCU will automatically release the dynamic PDCH when

the number of idle PDCH is enough.

2.2.12 Half Rate

I. Function Description

Increasing BSS capacity. With the half rate function, a TRX can provide a maximum of

16 half rate traffic channels (TCHs) and can simultaneously support a maximum of 16

MSs. This makes the BSS capacity almost doubled.

Raising call proceeding rate. With the half rate function, a half rate channel can be

assigned via the immediate assignment procedure. This raises the call proceeding rate.

Since the BSS system can provide more TCHs, there is no need to worry about

channel congestion even though the TCH is assigned during the immediate

assignment procedure.

II. Technical Description

1) Call procedure

The fundamental principle of the half rate function is that two logical half rate TCHs are

multiplexed in a timeslot of a physical TDMA frame as two logical channels. The

hardware provides a more advanced speech coding/decoding algorithm to make the

speech QoS on a half rate channel close to that on a full rate channel.

The BSC deals with the same call signaling procedures as before after the half rate

function is performed. However, the contents of some signaling messages may change

because a different channel type is selected. For example, that BSC allocates a half

rate TCH to MS may be reported to the MSC in an Assignment Complete or Handover

Complete message. The BSC shall select a speech version (in the case of speech

transmission) or data rate (in the case of data transmission) for the current call after

allocating a TCH. Due to the FTC capability limitation, the BSC does not support half

rate data services by default. The operator may enable the corresponding software

parameter switch (see "Parameter ") to enable the BSC to support the half rate data

services. The BSC always select the first data rate the MSC allows. For selecting a

speech version, a multi-module BSC takes into consideration the speech version the

MSC allows, the rate type of the current channel, the speech version supported by the

circuit pool of the A interface circuit of the current call and the capability of the FTC

bearing the current circuit. Then the BSC fetches the intersection of the above four sets.

It selects the latest one among the speech versions (if so) in the intersection. A





2-126

single-module BSC differs from a multi-module BSC in speech version selection. The

single-module BSC does not take the capability of the FTC bearing the A interface

circuit into consideration. Therefore, the A interface circuit configuration in a

single-module BSC should comply with the principle that the speech version setsupported by the circuit pool should be a subset of that supported by the FTC.

The dynamic switchover between a full rate TCH and a half rate TCH makes it possible

to optimize the TCH configuration according to the current capacity situation on the

existing resource basis. This lessens the possibility of TCH congestion and makes it no

longer a problem that the initially configured full rate TCHs and half rate TCHs cannot

satisfy the actual traffic requirement.

2) MSC channel rate selection policy

When the requested channel type is "Only select full rate channel" or "Only select half

rate channel ", only a channel at a fully matched rate can be allocated. When the

requested channel type is "Select full rate channel priority", a full rate TCH shall be

allocated if other conditions are satisfied and there is a full rate TCH in the cell. When

the requested channel type is "Select half rate channel priority ", a half rate TCH shall

be allocated if other conditions are satisfied and there is a half rate TCH in the cell. This

rigid channel allocation as per MSC's rate assignment is difficult to get the system

capacity and speech QoS into the optimum status. To break down this limitation,

Huawei introduces a BSC channel rate selection policy. The MSC channel rate

selection policy is still provided in order that the A interface interconnection test may

prove that channel allocation can be implemented as per MSC's assignment.

3) BSC channel rate selection policy

When the requested channel type is" Only select full rate channel " or " Only select half

rate channel ", only a channel at a fully matched rate can be allocated. When the

requested channel type is " Select full rate channel priority " or " Select half rate channel

priority ", a full rate TCH is preferred to guarantee the speech QoS if there are many idle

full rate TCHs and a half rate TCH is preferred to guarantee the system capacity if there

are few full rate TCHs. This is the basic principle of BSC channel rate selection. In

practice, a full rate TCH is preferred when the number of idle full rate TCHs > Idle Thrsh

for TCH/F Priori in the current cell and a half rate TCH is preferred when the number of idle full rate TCHs ≤ Idle Thrsh for TCH/F Priori in the current cell.

III. Parameter

1) Channel Management Parameter

"TCH Rate Adjust Allowed" in [Channel\Radio CH management ctrl. table].

"TCH Rate Adjust Traffic Thrsh" in [Channel\Radio CH management ctrl. table]

"MIN Recovery Time of TCH/H (s)" in [Channel\Radio CH management ctrl. table]

"Idle Thrsh for TCH/F Priori" in [Channel\Radio CH management ctrl. table]





2-127

2) Networking Parameter

The 34BIE has to be used to support the half rate function. Since the 34BIE applies a

different exchange mode from before, the following networking parameters are added

or modified.

"Connection mode" in [Local Office\BSC BIE description table].

"BIE networking configuration" in [Local Office\Site BIE config. table].

"Site ID 1 ~ Site ID 30" in [Local Office\BSC BIE active/stby. group table].

"Belong to BIE group No.1 ~ Belong to BIE group No.8" in [Site\Site description table].

"BSC BIE Port No " in [Local Office\Radio channel config. table].

"BSC BIE E1Timeslot No " in [Local Office\Radio channel config. table].

"BSC BIE port No " in [Local Office\LAPD semi-perm. connection table].

"BSC BIE E1Timeslot No "in [Local Office\LAPD semi-perm. connection table].

A [BSC BIE semi-perm. connection table] is added to support semi-perm. connection

establishment. It includes the following fields:

"Module No."in [Local Office\BSC BIE semi-perm. connection table]

"Trunk circuit No."in [Local Office\BSC BIE semi-perm. connection table]

"Transfer Rate "in [Local Office\BSC BIE semi-perm. connection table]

"BIE Port No "in [Local Office\BSC BIE semi-perm. connection table]

"E1 Timeslot No "in [Local Office\BSC BIE semi-perm. connection table]

2.2.13 E1 Ring Topology

I. Overview

E1 ring topology is a networking mode in which several sites are connected in a ring. Allthe sites work in the forward ring normally. In case the transmission ring is broken at

one point, the sites before the breakpoint still work in the forward ring while those after

the breakpoint are reinitialized and begin to work in the reverse ring. Compared with the

normal chain topology, E1 ring topology has an advantage that the transmission ring

can be automatically divided into two chains when it is broken at one point so that the

sites before and after the breakpoint can both still work normally. This enhances the

robustness of the system.





2-128


To be specific, the E1 ring topology has the following functions:

1) Automatic rotate in case of ring breakage

Normally all sites in a ring work in the forward ring as if they were in a normal chain. In

case the transmission ring is broken at one point, the sites after the breakpoint

automatically form a chain in the reverse direction and begin to work in the reverse ring

after reinitialization. If the transmission in the reverse ring is normal, the sites do not

automatically rotate back to the forward ring after the forward ring recovers. If the

transmission in the reverse ring is interrupted, the sites automatically rotate back to the

forward ring after the forward ring recovers. In the case of rotate (rotate-back), the BSC

reports a ring rotate (rotate-back) alarm to the alarm system to notify the operator to

examine and repair the transmission ring and so on.

2) Manual rotate/rotate-back

If they work normally in the reverse ring, the sites do not automatically rotate back to the

forward ring after the forward ring recovers from failure. The operator can forcedly

rotate them back to the forward ring. The operator can also manually rotate the sites to

the reverse ring in case the transmission quality in the forward ring is not good. The

manual rotate/rotate-back for a site is performed by specifying port 1/port 0 of the site

as a reset port in the site maintenance system during the site reset process. To realize

a manual ring rotate, the operator shall begin with the site at the first level in the reverse

ring to be rotated and then the one at the next level and so on. A site is successfully

rotated to the reverse direction and then reinitialized before the next site is rotated. To

realize a manual ring rotate-back, the operator shall begin with the site at the first level

in the forward ring to be rotated and then the one at the next level and so on.

3) Dynamic rotate parameter configuration and query

The operator may query the ring topology parameters and the current ring direction of a

site in the site maintenance system. The operator may also dynamically modify those

parameters in the data management system or auto data configuration system.

4) Dynamic data configuration

The BSC still supports dynamic data configuration in the E1 ring topology, such as

dynamic BTS adding/deletion, cell adding/deletion, TRX adding/deletion, etc. The

dynamic BTS adding must be implemented in the forward ring at the time when the

forward transmission is normal. All sites in a ring shall be reset after dynamic

adding/deletion of BTS, cell, TRX or modification of channel type in the ring or tributary.

5) Involved Technology





2-129

When it detects OML breakage, a site in a ring continuously tries to establish a chain via

port 0 or port 1 till it succeeds. The MPU prepares two sets of data i.e., forward ring data

and reverse ring data for each site. The BSC sends the forward ring data to a site when

a chain is established for the site in the forward ring. Otherwise, the BSC sends thereverse ring data to the site when a chain is established in the reverse ring. The site is

initialized and started after it receives the data.

III. Parameter

This parameter determined a group of BIE working under “Full Rate Ring Topology” or

“Half Rate Ring Topology”

Rotate parameters:

1) Auto rotate permit: It indicates whether a site is allowed to rotate automatically incase of transmission interruption. Its default value is Yes.

2) Waiting time before rotate: It indicates the time measured in seconds a site waits

before it rotates to the reverse direction in case of transmission interruption. Its value

range is 60~300. Its default value is 90.

3) Try rotating duration time: A site continuously tries to establish a chain via port 0 or

port 1 after it begins to rotate. It turns to the other port if it has not established a chain

via one port after the "Try rotating duration time". This time is measured in seconds.

The value range of this parameter is 60~300 and the default value is 90.

2.2.14 GSM900/GSM1800 Co-cell

I. Overview

With the sharp increase in MS quantity, it has become an inevitable tendency to

construct a dual band network. At present, there are three networking modes available

to the construction of dual band network: standalone MSC, shared-MSC but

standalone BSC, and shared-BSC. In any of the above networking modes, the inter-cell

handover and cell reselection will be inevitable, as will inevitably reduce the networkquality. A 900M/1800M co-cell is a networking mode in which the GSM900 and

DCS1800 TRXs coexist in the same cell. The best advantage of using the 900M/1800M

co-cell to construct a dual band network is that two frequency bands coexist in a cell

and that the 1800M frequency band becomes a natural extension of 900M frequency

band in this mode. This avoids cell reselection and inter-cell handover that are

inevitable in other networking modes.





2-130


A dual band MS can be freely handed over between the two frequency bands. Since

the 900M is characterized by less propagation loss and larger coverage, it is

recommended as the BCCH in this networking mode. The TRXs in the frequency band

the BCCH belongs to shall be in the underlaid subcell and those in the other frequency

band the BCCH does not belong to shall be in the overlaid subcell.

A 900M/1800M co-cell is based on a concentric platform. So does a two-timeslot

extended cell. However, the two types of cells have different application purposes: the

former is used for continuous coverage and scattered coverage in hot spots while the

latter for wide coverage. Therefore, a 900M/1800M co-cell repels a two-timeslot

extended cell. That is, a 900M/1800M co-cell cannot be a two-timeslot extended cell,

cannot be a 900M/1800M co-cell, and vice versa.

Note: The GSM900 in this document includes P-GSM, E-GSM and R-GSM.

1) 900M/1800M Co-cell Channel Allocation

Since a 900M/1800M co-cell is based on a concentric platform, the 900M/1800M

co-cell channel allocation shall comply with the concentric channel allocation strategy.

However, the BSC shall distinguish the MS frequency band capability before

performing channel allocation, for a single band MS may not support the frequency

band in the overlaid subcell. Therefore, for an MS supporting the frequency bands in

both overlaid and underlaid subcells, the channel allocation can be implemented as per

the concentric channel allocation strategy. For other MSs, only the channels in the

underlaid subcell shall be allocated. Details are given below:

Immediate assignment:

When immediate assignment is performed in a 900M/1800M co-cell, since no MS

information is available for reference, an underlaid-preferred channel allocation

strategy is adopted to guarantee the MS can most possibly initiate a call. That is, the

channel in the underlaid subcell shall be preferred.

Assignment:

When the MS classmark 3 is not obtained or when the MS classmark 3 indicates only

the frequency band of the underlaid subcell is supported, only the channel in the

frequency band of the underlaid subcell can be allocated no matter how the Assign

Optimum Layer is configured. When the MS classmark 3 indicates both frequency

bands are supported, the channel can be allocated according to the Assign Optimum

Layer configuration.

2) Intra-BSC incoming cell HO:


the frequency band of the underlaid subcell is supported, only the channel in thefrequency band of the underlaid subcell can be allocated no matter how the Pref.





2-131

Subcell in HO of Intra-BSC is configured. When the MS classmark 3 indicates both

frequency bands are supported, the channel can be allocated according to the Pref.

Subcell in HO of Intra-BSC configuration.

3) Inter-BSC incoming cell HO:


the frequency band of the underlaid subcell is supported, only the channel in the

frequency band of the underlaid subcell can be allocated no matter how the Pref.

Subcell in HO of Inter-BSC is configured. When the MS classmark 3 indicates both

frequency bands are supported, the channel can be allocated according to the Pref.

Subcell in HO of Inter-BSC configuration.

4) 900M/1800M Co-cell Handover

Since a 900M/1800M co-cell is based on a concentric platform, the 900M/1800M

co-cell handover shall comply with the normal concentric handover strategy.

III. Parameters:

"Cell system type " in [Local Office\BSC Cell Table]

" HW-IUO property" in [Site\Carrier Configuration Table]

Table name: [Handover\Concentric Cell HO Table]

Parameter: All data in [Concentric Cell HO Table] are applicable to a 900M/1800M

co-cell.

2.2.15 Multi-MNC

I. Overview

The multi Mobile Network Code (multi-MNC) function allows the operator to configure

the cells which have different MNC in one BSC.


1) System message processing

The system sends different system messages including different MNCs respectively to

the multi-MNC cell and the normal cell, and then an MS can display the mobile network

name it subscribes to as per the system message. (The BSC contains the same Cell

Global Identification (CGI) for a cell as the MSC does if the MSC supports a multi-MNC

cell. In this case, the BSC need not use the multi-MNC function parameter).

The multi-MNC function is applied when two or more network operators are integrated

or when some small operators rent equipment from a large operator. See Figure 2-42.





2-132

A-MSC

A-BSS

Cell 1: 460 12 1850 0001

Cell 2: 460 34 1810 0002

T

R

X

1

T

R

X

2

MS Display: 12 MS Display: 34

B-MSC

Operator A: MNC = 12

Operator B: MNC = 34

Figure 2-42 Multi-MNC diagram

In this document, a multi-MNC cell is a cell whose CGI includes a MNC different from

the MNC configured in the [Local Office Information Table].

2) Handover strategy

System provides flexible handover control means. In different situations, mobile

phones can be handed over to a cell with the same MNC or a cell with a different MNC.

There are seven handover control means provided: normal handover, only the

handover to a cell with the same MNC allowed, a cell with the same MNC first, a better

cell with the same MNC first, a better cell with a different MNC first, a cell with a different

MNC first, and only the handover to a cell with a different MNC allowed.

Select a suitable multi-MNC handover control type according to the actual situation.

The following will give an introduction to the control strategy and the possible

applicable situation of each handover type.

In the situations where multiple MNCs are used, configure "Multi-MNC handover

judgement allowed" with "Y". "Multi-MNC handover type" can be configured with the

expected types according to the actual requirements.

The control strategy and applicable situation of each handover type are as follows:

a) Normal handover

Control strategy: Handover to a cell with better QoS (considering all such factors as

level, quality, cell level, load, whether to share the same BSC (MSC), etc.).





2-133

Applicable situation: The situations where mobile phones are expected to be handed

over to a cell with better QoS. Not considering whether the object cell has the same

MNC or a different MNC.

b) Only the handover to a cell with the same MNC allowed

Control strategy: Only the handover to a cell with the same MNC is allowed, including

the service cell.

Applicable situation: The situations where mobile phones are expected to be handed

over to a cell in the same network.

c) A cell with the same MNC first

Control strategy: As long as the adjacent cell has the same MNC with the service cell

and the received level is higher than "minimal downlink power of candidate handover cell" of this adjacent cell, mobile phones will be handed over to this adjacent cell,

including the service cell.

Applicable situation: As long as a cell in the same network can provide normal

services, mobile phones are expected to be handed over to this cell. When mobile

phones cannot be handed over to a cell in the same network (for example, no signal

can be detected in a cell with the same MNC), they can be handed over to a cell with a

different MNC.

d) A better cell with the same MNC first

Control strategy: If the adjacent cell and the service cell have the same MNC and the

received level of the adjacent cell is higher than the inter-layer handover threshold of

this adjacent cell, mobile phones can be handed over to this adjacent cell, including the

service cell.

Applicable situation: When a cell in the same network can provide good services (i.e.,

higher than the inter-layer handover threshold), mobile phones are expected to be

handed over to a cell in the same network. If no cell that can provide good services is

available in the same network, mobile phones will be handed over to a cell with the

better QoS, regardless of whether the object cell with the same MNC or a differentMNC.

e) A better cell with a different MNC first

Control strategy: If the adjacent cell has a different MNC from the service cell and the

received level of the adjacent cell is higher than the inter-layer handover threshold,

mobile phones will be handed over to this adjacent cell.

Applicable situation: When a cell with a different MNC from the service cell can

provide good services (i.e., higher than the inter-layer handover threshold), mobile

phones are expected to be handed over to this cell. If no cell that can provide good





2-134

services is available in other networks, mobile phones will be handed over to a cell with

the better QoS. If there are only a few cell channels in this network but the coverage is

satisfied, this control strategy can be selected in case of congestion and if traffic

sharing is allowed in other networks.

f) A cell with a different MNC first

Control strategy: As long as the adjacent cell has a different MNC from the service cell

and the received level of the adjacent cell is higher than "minimal downlink power of

candidate handover cell", mobile phones will be handed over to this adjacent cell.

Applicable situation: As long as a cell in another network can provide normal services,

mobile phones are expected to be handed over to this cell. When mobile phones

cannot be handed over to a cell in another network (for example, no signal can be

detected in this cell), mobile phones can be handed over to a cell with the same MNC.If there are only a few cell channels in this network but the coverage is satisfied, this

control strategy can be selected in case of congestion and if traffic sharing is allowed in

other networks.

g) Only the handover to a cell with a different MNC allowed

Control strategy: Only the handover to a cell with a different MNC is allowed.

Applicable situation: Mobile phones are expected to be handed over only to a cell in

another network.

3) Application note

a) The CGI allocated by the MSC to a normal cell should include a LAC different from

the LAC in the CGI the MSC allocates to a multi-MNC cell.

b) The MNC in the [Local Office Information Table] at the BSC side should be the same

as the MNC configured at the MSC side. At the BSC side, the cell CGI is configured as

the CGI over the Abis interface of the local BSC and the external cell CGI as the CGI

over the Abis interface of the peer BSC. There should be only one difference between

the multi-MNC cell CGI configured at the BSC side and the corresponding CGI at the

MSC side i.e., different MNCs.

c) For a BSC with the multi-MNC function, inter-BSC handover can be implemented

only when the peer BSC is also designed as multi-MNC function supportable.

d) The multi-MNC cell cannot support GPRS services presently and the normal cell

can.

III. Parameter

"Multi-MNC HO Allowed " in [Handover/ Handover Control Data]

"Multi-MNC HO Type" in [Handover/ Handover Control Data]





2-135

2.2.16 E-GSM/R-GSM

I. Overview

Along with the development of GSM in a large scale, the frequency resource becomes

more and more insufficient and bottlenecks the further development of GSM. The

current solution is to introduce new frequency bands. The introduction of E-GSM and

R-GSM extended bands plays an important role in solving the shortage of frequency

resource.


According to GSM 05.05 (version 8.5.0), there are four frequency bands:

1) GSM900 base band, P-GSM:

The working frequencies of the GSM900 baseband are:

890 ~ 915MHz: For mobile phone sending and BTS receiving.

935 ~ 960MHz: For BTS sending and mobile phone receiving.

2) GSM900 extended band, E-GSM (including GSM900 base band):

The working frequencies of GSM900 extended band are:



3) GSM900 railway band, R-GSM (including GSM900 base band and GSM900

extended band):

The working frequencies of R-GSM900 are:



4) DCS1800 band:

The working frequencies of DCS1800 band are:



The internal between frequencies is 200kHz.

The relations between frequency points and absolute frequencies are as follows. n:

frequency point. Fl(n): uplink frequency corresponding to n. Fu(n): downlink frequency

corresponding to n. Unit: MHz.





2-136

Table 2-17 Table 1 Relations between frequency points and absolute frequency

P-GSM 900 Fl(n) = 890 + 0.2*n 1<= n <=124 Fu(n) = Fl(n) + 45

E-GSM 900 Fl(n) = 890 + 0.2*n 0<= n <=124 Fu(n) = Fl(n) + 45

Fl(n) = 890 + 0.2*(n-1024) 975<= n <=1023

R-GSM 900 Fl(n) = 890 + 0.2*n 0<= n <=124 Fu(n) = Fl(n) + 45

Fl(n) = 890 + 0.2*(n-1024) 955<= n <= 1023

DCS 1800 Fl(n) = 1710.2 + 0.2*(n-512) 512<= n <= 885 Fu(n) = Fl(n) + 95

The newly introduced E-GSM 900 band and R-GSM 900 band belong to the same

band with P-GSM. However, the frequency points are not continuous. Therefore, the

E-GSM extended band and R-GSM extended band are introduced. The E-GSM

extended band refers to the E-GSM band excluding the P-GSM band provided in the

Protocol. The R-GSM extended and refers to the R-GSM band excluding the E-GSM

provided in the Protocol.

Huawei BSC can support four frequency bands: P-GSM, E-GSM, R-GSM, and

DCS1800.

Table 2-18 Table 2 Relations between frequency bands and frequency points

Frequency band

P-GSM

E-GSM R-GSM DCS1800 E-GSMextendedband

R-GSMextendedband

Frequency point

1~124 0~124,975~1023

0~124,955~1023

512~885 0, 975~1023 955~974

Channel assignment technology of E-GSM\R-GSM

For the cell configured with frequency points of the E-GSM extended band or the

R-GSM extended band, the channel assignment technology can adopt different

assignment strategies in different situations after fully analyzing the frequency bandsupport capabilities of the mobile phone and channels.

I generation algorithm of channel assignment: Before assignment, the system has

obtained the classmark of the mobile phone. According to it, the system can work out

the support capability of each channel for this mobile phone. And then the system can

assign a channel from all channels supporting this mobile phone conforming to the

polling strategy. For example, if a mobile phone supports E-GSM, the channel to be

assigned can be a channel of P-GSM or E-GSM extended band.

II generation algorithm of channel assignment: Before assignment, the system has

obtained the classmark of the mobile phone. According to it, the system can work out





2-137

the support capability of each channel for this mobile phone. In all channels supporting

this mobile phone, the system gives the priority to the channel of the band outside the

band intersection. For example, if the mobile phone supports E-GSM and the channels

respectively support the P-GSM band and the E-GSM extended band. The channel of the E-GSM extended and will be assigned first. The band intersection P-SGM will be

reserved for other mobile phones with poor support capability.

For the immediate assignment, the system assigns the channel for the mobile phone

according to the frequency band support capability of the host BCCH.

III. Parameters

"Effective frequency points: 1~64" in [Carrier configuration table]

2.3 GPRS Function

The GPRS functions supported by BSS include:

Radio link management and radio resources management.

Access control over the MS.

Provision of routes for the transfer of packet data.

The radio link management function involves mainly establishment, maintenance and

release of the radio links. The radio resources management function involves mainly

coding/decoding of radio packet channels, configuration of radio packet channels,

multiplexing of radio channels, switchover of radio channels between the circuit

switched traffic channel and the packet switched traffic channel and allocation of

channels to the MS.

The access control function serves primarily to solve the issue of channel contention

and allocate radio resources to the MS according to the requested QoS.

The routes serve to transfer the uplink data to Serving GPRS Support Node (SGSN)

properly and receive the downlink data from the SGSN.

2.3.1 Supported Packet System Information

I. Overview

The packet system information broadcast in the cell serves mainly for the MS to access

the network. If the cell supports GPRS, the System Information 13 (SI13) shall be

added to the BCCH, if not, SI13 will not be broadcast. The cell can either be configured

with the PBCCH channel or without it. This will be notified to the MS via SI13. The main





2-138

message broadcast on the PBCCH are the dedicated packet system information of

GPRS.


There are following types of packet system information:

1) Packet System Information Type 1 (PSI1)

This information is sent by the network on the PBCCH or PACCH, giving information for

Cell selection, for control of the PRACH, for description of the control channel(s) and

optional global power control parameters. This information shall not be segmented

across more than one RLC/MAC control block.


This information is sent by the network on PBCCH and PACCH, giving information of reference frequency lists, cell allocation, GPRS mobile allocations and PCCCH

descriptions being used in the cell. PSI2 also contains Non-GPRS cell options

applicable for non-packet access. This message shall not be segmented across more

than one RLC/MAC control block.


This information is sent by the network on the PBCCH or PACCH, giving information of

the BCCH allocation (BA_GPRS) in the neighboring cells and cell selection parameters

for serving cell and non-serving cells. This information shall not be segmented across

more than one RLC/MAC control block.

4) Packet System Information Type 3 bis (PSI3bis)

This information is sent by the network on the PBCCH and PACCH giving information of

the BCCH allocation in the neighboring cells and cell selection parameters for

non-serving cells. This information shall not be segmented across more than one

RLC/MAC control block. If not all information fits into one instance of the PSI3bis, the

PSI3bis can be repeated.


This information is optionally sent by the network on the PBCCH and PACCH giving

information directing the mobile station to make measurements on a list of serving cell

PDCHs, during the idle frame of those PDCHs. This information shall not be

segmented across more than one RLC/MAC control block.


This optional information is sent by the network on the PBCCH giving information for

measurement reporting and network controlled cell reselection. If not all information fits

into one information, the remaining information will be sent in other instances of the

PSI5. The information is sent on PBCCH only if so indicated in PSI1.






2-139

This information may be broadcast by the network on the PACCH. The information

provides the mobile station with GPRS cell specific access-related information. The

information in this information shall be the same as provided in the PSI13 on BCCH.

PSI1~PSI4 can either be broadcast on the PBCCH or sent on the PACCH. PSI5 can be

broadcast only on the PBCCH. PSI13 can be sent only on the PACCH. When there is

PBCCH in the cell, no PSI13 will be transferred on the PACCH, but PSI1 will be

broadcast periodically on it. When there is no PBCCH in the cell, only the PSI13 will be

broadcast periodically on the PACCH.

M900/M1800 BSS can transfer all the GPRS-related packet system information,

provide such functions as controlled re-transmission, high-speed re-transmission and

low-speed retransmission, and control the transfer of packet system information on the

PACCH based on the configuration of the PBCCH/PCCCH in a cell.

III. Parameter

At present, the PBCCH and PCCCH are not configured. Therefore, PSI13 is usually

broadcasted on PACCCH.

The configuration of PSI13 is realized with the command pcu add gprs. The following

parameters are involved.

1) NMO

Description: network operation mode.

Value range: 0~3. 0 - network operation mode I, 1 - network operation mode II, 2 -

network operation mode III, 3 – reserved.

Default: At present, the GPRS network usually has neither Gs interface nor PCCCH

configured. Therefore, this parameter is usually set as "1".

2) T3168

Description: T3168 timer overtime value. The duration for MS to wait for the packet

uplink assignment message.

Value range: 500 ms, 1000 ms,↑, 4000 ms

Default: 1000ms

T3192

Description: T3192 timer overtime value. The duration for the MS to wait for TBF

release after receiving the last data block.

Value range: 500 ms, 1000 ms, 1500 ms, 0 ms, 80 ms, 120 ms, 160 ms and 200 ms

Default: 500 ms

3) DRXTimerMax





2-140

Description: maximum duration of non-DRX. The maximum duration for executing

non-DRX mode when MS is switching from packet transfer mode to packet idle mode.

Value range: no – enter DRX mode immediately, 1 s – 1 s, 2 s – 2 s, ↑64 s – 64 s

Default: 4 s, i.e., enter DRX mode after 4 seconds

4) AccBurst

Description: the access burst format used by MS in PRACH, PTCCH/U or packet

control acknowledge message.

Value range: 8 bit, 11bit

Default: Some MSs do not support 11 bit access burst type, so it is recommended to

set it as "8 bit".

5) CtrlAckType

Description: control acknowledge message type. The format adopted by MS in the

control acknowledge message.

Value range: 0, 1. 0 – 4 access burst used (TA can be obtained without sending

"polling" message), 1 - RLC/MAC control block used (TA can be obtained only after

sending "polling" message).

Default: 0

6) BsCvMax

Description: maximum value of MS countdown timer

Value range: 0~15

Default: 4

7) PanDec

Description: PAN_DEC used by MS N3102 counter. When MS T3182 times out,

N3102 will reduce the value of PAN_DEC.

Value range: 0~7, nouse

Default: 3

8) PanInc

Description: PAN_INC used by MS N3102 counter. When MS activates T3182 and

receives the corresponding acknowledge message from packet uplink, N3102 will

increase the value of PAN_DEC.

Value range: 0~7, nouse

Default: 3

9) PanMax





2-141

Description: PAN_MAX of MS N3102 counter, i.e. the maximum value of N3102.

Value range: 4, 8, ↑ 32, nouse

Default: 12

2.3.2 Supported GPRS MS Modes

I. GPRS MS types

The integrated GPRS MS consists of ME and SIM. ME includes MT and TE. GPRS MS

is divided into three classes:

1) Class A GPRS MS

Class A GPRS MS can simultaneously be connected to both GSM and GPRS networksand activated in both networks, monitor information of each system and start them

simultaneously, and provide GPRS service and GSM circuit-switched service

simultaneously, including short message service. Class A MS subscriber can

initiate/receive call in both services and handover automatically between packet data

service and circuit service.

2) Class B GPRS MS

Class B GPRS MS can be connected to both GSM and GPRS networks simultaneously.

It can be used in GPRS packet service and GSM circuit-switched service separately but

not simultaneously, i. e. in a certain moment, it uses either circuit-switched service or packet-switched service. Class B MS also can automatically handover.

3) Class C GPRS MS

At a certain moment, Class C MS can be connected only GSM network or GPRS If it

supports both packet-switched and circuit-switched services, the service handover

should be completed manually. It cannot use both services simultaneously.

II. GPRS MS multislot hierarchy

GPRS system can use the MAC layer function to provide a subscriber with the multislot

mode. However, the corresponding MS should be capable to support this mode. The

classification of MS under different multislot modes is listed as follows.





2-142

Table 2-19 Classification of MS under different multislot modes

Classification Max. number of time slots Min. number of time slots MS type

Rx Tx Sum Tta Ttb Tra Trb

1 1 1 2 3 2 4 2 1

2 2 1 3 3 2 3 1 1

3 2 2 3 3 2 3 1 1

4 3 1 4 3 1 3 1 1

5 2 2 4 3 1 3 1 1

6 3 2 4 3 1 3 1 1

7 3 3 4 3 1 3 1 1

8 4 1 5 3 1 2 1 1

9 3 2 5 3 1 2 1 1

10 4 2 5 3 1 2 1 1

11 4 3 5 3 1 2 1 1

12 4 4 5 2 1 2 1 1

13 3 3 NA NA a 3 a 2

14 4 4 NA NA a 3 a 2

15 5 5 NA NA a 3 a 2

16 6 6 NA NA a 2 a 2

17 7 7 NA NA a 1 0 2

18 8 8 NA NA 0 0 0 2

19 6 2 NA 3 b 2 c 1

20 6 3 NA 3 b 2 c 1

21 6 4 NA 3 b 2 c 1

22 6 4 NA 2 b 2 c 123 6 6 NA 2 b 2 c 1

24 8 2 NA 3 b 2 c 1

25 8 3 NA 3 b 2 c 1

26 8 4 NA 3 b 2 c 1

27 8 4 NA 2 b 2 c 1

28 8 6 NA 2 b 2 c 1

29 8 8 NA 2 b 2 c 1

a=1 indicates that FH is adopted; a=0 for not adopting FH; b=1 indicates that FH being adopted or MS changes





2-143

frequency when it changes from receiving to transmitting; b=0 indicates that FH is not adopted and MS does not

change frequency when it changes from receiving to transmitting; c=1indicates that FH is adopted and MS changes

frequency when it changes from transmitting to receiving; c=0indicates that FH is not adopted and MS does not

change frequency when it changes from transmitting to receiving .Rx indicates the maximum number of time slots used in an MS downlink in a TDMA frame. MS should support

configurations of time slot numbers indicated by all integers from 0 to RX. The receiving time slots can be

discontinuous by time. For the type 1 MS, its receiving time slots will be distributed in a receiving window with a size of

RX. There is no transmitting time slot between receiving time slots in the TDMA frame.

Tx indicates the maximum number of time slots used in an MS uplink in a TDMA frame. MS should support

configurations of time slot numbers indicated by all integers from 0 to Tx The transmitting time slots can be

discontinuous by time. For the type 1 MS, its transmitting time slots will be distributed in a transmitting window with a

size of RX. There is no receiving time slot between transmitting time slots in the TDMA frame.

Sum indicates the sum of all available time slots that can be used by MS in each TDMA frame. 1 ≤Rx + Tx≤Sum

Tta relates to the time needed for the MS to perform adjacent cell signal level measurement and get ready to transmit.

Ttb relates to the time needed for the MS to get ready to transmit.

Tra relates to the time needed for the MS to perform adjacent cell signal level measurement and get ready to receive.

Trb relates to the time needed for the MS to get ready to receive.

Huawei’ BSS system supports

MS of Type B

MS of Type C

MS with multi-slot capability ranging 1~12.

2.3.3 Supported RLC Modes

I. RLC acknowledged mode

Under RCL acknowledged mode, the transmission of RLC data block adopts

retransmission method. The transmitting party numbers RLC data block through block

sequence number (BSN), which can be used for retransmission and recombination.

The receiving party can transmit “PACKET ACK/NACK” message to request retransmit

RLC data block.

II. RLC non-acknowledged mode

The transmission of RLC data block under RLC non-acknowledged mode does not

support retransmission. But in the process of releasing uplink TBF, the last transmitted

uplink block might be retransmitted. The BSN of RLC data block header numbers the

RLC block data for recombination. The receiving party sends “PACKET ACK/NACK” to

transmit other necessary control signaling.

Huawei GPRS BSS supports RLC acknowledged mode and non-acknowledged mode.





2-144

2.3.4 Supported Channel Coding Scheme

PDTCH defines four coding schemes: CS-1~CS-4. Except for PRACH and TPCCH/U,

CS-1 is usually used for other packet control channels. For PRACH access burst,defines two coding schemes. All coding schemes are forced for MS. Only CS-1 is

forced for network.

I. Channel coding:

Channel coding of PDTCH

The radio block bearing RLC data block can use 4 types of coding schemes.

Parameters of each coding scheme are shown in Table 2-20.

Table 2-20 Coding parameter of coding schemes

Channel coding scheme RLC/MAC data block size (octets) Rate kbit/s

CS-1 23 9. 05

CS-2 33 13. 4

CS-3 39 15. 6

CS-4 53 21. 4

Channel coding of PACCH, PBCCH, PAGCH, PPCH, PNCH and PTCCH

PACCH, PBCCH, PAGCH, PPCH, PNCH and downlink PTCCH adopt CS-1, uplink

PTCCH adopts the same coding scheme as PRACH.

Channel coding of PRACH

There are two types of packet access burst in PRACH: 8bit and 11bit packet access

burst.

8bit packet access burst bears 8 information bits. The same channel coding is adopted

for uplink packet access burst and random access burst. The coding scheme of 11-bit

packet access burst bears 11 information bit.

Huawei BSS supports all these four CS and dynamically handovers between them

according to radio transmission quality (RLC block retransmission rate of uplink

/downlink TBF).

II. Dynamic Additional Sub-TS

Under CS-1/CS-2, BSS is also based on the 16kbit/s link at the G-Abis interface. When

CS-3 and CS-4 is adopted, the rate of a PDCH is 15.6 kbit/s and 21.4 kbit/s, therefore,

when mapping the radio channels to the terrestrial channels, a PDCH is mapped to two

16 kbit/s links. However, the coding scheme adopted by PDCH is adjusted according tothe change of the radio transmission environment of the MS that occupies it. Mapping a





2-145

PDCH permanently onto two 16kbit/s links will greatly decrease the multiplexing ratio of

the G-Abis interface, and thus greatly reduce the utilization ratio of the G-Abis interface

transmission equipment.

With dynamic additional sub-TS allocation, M900/M1800 GPRS BSS can resolve the

transmission issue of CS-3 and CS-4 on the G-Abis interface perfectly. The dynamic

attached sub-slot technology is to statically allocate a main 16 kbit/s sub-timeslot and

dynamically allocate a attached 16 kbit/s sub-timeslot at the G-Abis interface for the

CS-3/CS-4 PDCH. With the dynamic additional sub-TS technology, it is not necessary

for GPRS BSS to upgrade the hardware of BTS, BSC or PCU for supporting CS-3 and

CS-4. In addition, in its support for CS-3 and CS-4, the multiplexing ratio of the G-Abis

interface is greatly improved, thus saving investments on the G-Abis interface

transmission equipment.

The dynamic additional sub-TS technology used by M900/M1800 GPRS BSS displays

the following features:

Any idle Sub-TS of the G-Abis interface can be used as additional sub-TS, so that

each has maximum utilization.

Within a same site address, the additional sub-TSs can be dynamically attached to

various main TSs to enhance the utilization ratio of this sub-TSs according to

statistical multiplexing rules.

The locations of the additional 16kbit/s sub-TS are relatively flexible. They do not

have to be adjacent to the main 16kbit/s sub-TS.

It packs and unpacks the data packets through software to avoid hardware

upgrading.

III. Parameter

M900/M1800 BSS supports four types of coding schemes: CS-1~CS-4. The

configuration related to channel coding schemes is realized with the command pcu

add cspara. The following parameters are involved.

1) UpFixCs

Description: CS fixedly adopted for uplink.

Value range: cs1, cs2, cs3, cs4, unfixed

Default: cs2

2) UpDefaultCs

Description: Default CS adopted for uplink. If the uplink configured to adjust CS type

dynamically, then the CS of the first TBF to transfer can be set by this parameter, the

CS of other TBF is dynamically adjusted according to the signal transmission quality.

Value range: cs1, cs2, cs3, cs4.

Default: cs2





2-146

3) UpThdCs1Cs2

Description: The resend rate conversion threshold, i.e. when the resend rate of the

uplink TBF is smaller than or equals to this value, the coding scheme of it changes from

CS-1 to CS-2.

Value range: 0~100

Default: 5

4) UpThdCs2Cs1

Description: Uplink TBF retransmission rate conversion threshold from CS-2 to CS-1,

i.e. when the uplink TBF retransmission rate is larger than or equals to this value, the

uplink TBF coding scheme changes from CS-2 to CS-1.

Value range: 0~100

Default: 10

5) UpThdCs2Cs3


i.e. when the uplink TBF retransmission rate is smaller than or equals to this value, the


Value range: 0~100

Default: 10

6) UpThdCs3Cs2




Value range: 0~100

Default: 20

7) UpThdCs3Cs4

Description: Uplink TBF retransmission rate conversion threshold from CS-3 to CS-4,i.e. when the uplink TBF retransmission rate is smaller than or equals to this value, the


Value range: 0~100

Default: 10

8) UpThdCs4Cs3








2-147

Value range: 0~100

Default: 20

9) DnFixCs

Description: Fixedly adopted CS for downlink.

Value range: cs1, cs2, cs3, cs4, unfixed.

Default: unfixed

10) DnDefaultCs

Description: Default CS adopted for downlink. If the downlink dynamically adjusts CS,

then the CS of the first TBF can be set by this parameter, the CS type of other TBF is

dynamically adjusted according to the signal transmission quality.

Value range: cs1, cs2, cs3, cs4.

Default: cs2

11) DnThdCs1Cs2

Description: Downlink TBF retransmission rate conversion threshold from CS-1 to

CS-2, i.e. when the downlink TBF retransmission rate is smaller than or equals to this

value, the downlink TBF coding scheme changes from CS-1 to CS-2.

Value range: 0~100

Default: 5

12) DnThdCs2Cs1


CS-1, i.e. when the downlink TBF retransmission rate is larger than or equals to this


Value range: 0~100

Default: 10

13) DnThdCs3Cs2




Value range: 0~100

Default: 20

14) DnThdCs3Cs4


CS-4, i.e. when the downlink TBF retransmission rate is smaller than or equals to this






2-148

Value range: 0~100

Default: 10

15) DnThdCs4Cs3




Value range: 0~100

Default: 20

16) MaxFixCs34Pdch

Description: The fixed maximum number of PDCH supporting CS-3/CS-4 in this cell.

Value range: 0~255

Default: none

2.3.5 Supported Network Control Modes

During the cell reselection required by network, the network requires MS to send MR so

as to control the cell reselection. Here three network control modes are defined.

NC0: MS controlled cell re-selection, no measurement reporting.

NC1: MS controlled cell re-selection, MS sends measurement reports. NC2: Network controlled cell re-selection, MS sends measurement reports.

Huawei BSS supports NC0.

I. Parameter

Network control mode parameter is configured with the command pcu add relatedinfo

in PCU.

Parameter : NCO

Description: Network control mode.

Value range: nc0, nc1, nc2

Default: Currently fixedly set as "nc0", meaning MS controlled cell re-selection, no

measurement reporting

2.3.6 Supported Network Operation Mode

GPRS network defines three network operation modes in order to uniformly coordinate

the pagings of circuit-switched and packet-switched service.





2-149

I. Network operation mode I

The network sends a CS paging message for a GPRS-attached MS, either on the same

channel as the GPRS paging channel (i.e., the packet paging channel or the CCCH

paging channel), or on a GPRS traffic channel. This means that the MS needs only to

monitor one paging channel, and that it receives CS paging messages on the packet

data channel when it has been assigned a packet data channel.

II. Network operation mode II

The network sends a CS paging message for a GPRS-attached MS on the CCCH

paging channel, and this channel is also used for GPRS paging. This means that the

MS needs only to monitor the CCCH paging channel, but that CS paging continues on

this paging channel even if the MS has been assigned a packet data channel.

III. Network operation mode III

The network sends a CS paging message for a GPRS-attached MS on the CCCH

paging channel, and sends a GPRS paging message on either the packet paging

channel (if allocated in the cell) or on the CCCH paging channel. This means that an

MS that wants to receive pages for both circuit-switched and packet-switched services

shall monitor both paging channels if the packet paging channel is allocated in the cell.

No paging co-ordination is performed by the network.

Table 2-21 shows channels that deliver circuit paging message and packet paging

message under various network modes.

Table 2-21 Network operation mode

Network operationmode

Circuit pagingchannel

GPRS pagingchannel

Pagingcoordination

Packet PagingChannel

Packet PagingChannel

CCCH PagingChannel

CCCH PagingChannel

I

Packet Data Channel Not Applicable

Yes

IICCCH PagingChannel

CCCH PagingChannel

No

CCCH PagingChannel

CCCH Packet PagingChannel

IIICCCH PagingChannel

CCCH PagingChannel

No

Currently, Huawei GPRS BSS system supports three network operation modes.





2-150

2.3.7 Supported QoS

GPRS provides the subscriber with negotiable QoS configuration. GPRS QoS has five

basic attributes:

Precedence class;

Delay class;

Reliability class;

Peak throughput class

Mean throughput class.

Each attribute has multiple values available. The combination of different priorities

enables the system to support various applications with different QoSs required.

During the negotiation of QoS profile, MS can apply a value for every QoS attribute,

including the default value stored in HLR and used to create new account. Network also

needs to negotiate a priority for every attribute so that it can keep consistent with

effective GPRS resources. Network always provides adequate resource to support the

negotiated QoS profiles. RLC/MAC layer supports four radio priority levels, and

whether the cause for the uplink access is user data or signaling message transmission.

This information is used by the BSS to determine the radio access precedence and the

service precedence. The radio priority levels to be used for transmission of MO SMS

shall be determined by the SGSN and delivered to the MS in the Attach Accept

message. The radio priority level to be used for user data transmission shall be

determined by the SGSN based on the negotiated QoS profile and shall be delivered tothe MS during the PDP Context Activation and PDP Context Modification procedures.

Huawei BSS can satisfy MS QoS requirements as much as possible according to the

state of current radio resources.

2.3.8 Supported Assignment

When network side or MS requests to establish TBF for data transmission, GSM/GPRS

network can assign some channel resources for data transmission or refuse the

request according the multi-slot capability of MS and network resources state.

Network can assign TBF resource from CCCH, PACCH or PCCCH. According to the

direction of TBF data transmission, assignment can be divided into uplink and downlink

assignment. When channel resource is in short or for other causes, network can refuse

the request to establish TBF.

When MS requests to establish TBF for uplink data transmission, network sends

immediate assignment message on CCCH or packet uplink assignment message on

PCCCH to assign channel resource. When MS requests to establish uplink TBF during

its downlink TBF data transmission, network assigns packet uplink channel for MS on

PACCH. MS transmits data according to assigned channel resource.





2-151

When network requests to establish TBF for downlink data transmission, network can

send immediate assignment message on CCCH or packet assignment message on

PCCCH for MS to assign downlink channel resource. When MS requests to establish

downlink TBF during its uplink TBF data transmission, network can assign packetuplink channel for MS through PACCH. MS transmits data according to assigned

channel resource.

Network assigns resource on different channels according to CCCH or PCCCH

configuration. Meanwhile, network can perform different assignments such as single

block assignment and packet resource assignment according to different access

requests, such as two phases, one phase, and single block request.

M900/M1800 BSS supports:

Packet uplink resource assignment on PACCH

Packet downlink resource assignment on PACCH

Uplink immediate assignment for TBF establishment on CCCH.

Downlink immediate assignment for TBF establishment on CCCH.

2.3.9 Supported Paging

In GPRS/GSM system, paging includes packet paging and circuit paging.

I. Packet paging

When there are downlink data that shall be sent to the MS, SGSN needs to initiate a

packet paging call so as to locate MS accurately. The paging request message

originated by the SGSN is sent through Gb interface to PCU, which converts it into the

packet paging request of the air interface (Um interface) before sending. If the PCCCH

channel is configured for the BSS system, the message will be sent on the PPCH

directly. If PCCCH is not configured for the system, PCU will send the message via the

Pb interface to the BSC, which sends it on the PCH.

After receiving packet paging message, MS will initialize the process of uplink TBF

establishment, and then sends the paging response packet in data form to PCU via the

air interface. PCU forwards the packet to SGSN. After receiving the paging response,

SGSN is ready to transmit downlink data.

II. Paging coordination

When a call reaches the MSC where the subscriber is located, the MSC sends a paging

message to all cells in that location area according to the registered location area of

MS.

If there is Gs interface between SGSN and MSC, GPRS/GSM system operates in

network operation mode I and the paging service of GSM service can be sent through





2-152

GPRS packet channel. In other words, if an MS is GPRS-attached, its circuit paging go

from MSC to SGSN and then to PCU through the Gs and Gb interfaces, and PCU will

determine on which channel to transmit the paging.

In network operation mode I, if this MS has been assign with packet dedicated channel,

then the circuit paging is sent on PACCH; if this MS has not assigned with packet

dedicated channel and the system has not been configured with PCCCH, PCU

forwards the paging message to BSC through Pb interface. Then BSC sends this

message on PCH.

If there is no Gs interface between SGSN and MSC, GPRS/GSM system can operate

only in network operation mode II and mode III. In this case, the system sends circuit

paging on CCCH.

After receiving the circuit paging message, MS accesses the RACH and starts thecircuit connection setup process. If MS is currently handling GPRS service, MS will

initiate GPRS Suspend process to suspend GPRS service and it will not recover GPRS

service until circuit connection is released.

M900/M1800 BSS supports above-mentioned packet paging and paging coordination.

2.3.10 Timing Advance

Timing advance (TA) procedure is used to extract the correct TA value so that MS can

transmit radio block on uplink. TA includes tow parts:

I. Initial TA estimation

Initial TA estimation is based on a single access burst bearing packet channel. Request,

packet uplink assignment or packet downlink assignment message, estimate TA and

send to MS. MS uses this value for uplink transmission till a new value is provided.

II. Continuous TA update

The MS in packet transmission mode needs continuous TA update. TA update is born

by PTCCH assigned to MS. Uplink packet transmission uses packet uplink assignment

message to assign TA index (TAI) and PTCCH to MS. And downlink packet

transmission uses packet downlink assignment message to assign TA index (TAI) and

PTCCH to MSTAI specifies the PTCCH sub-channel for MS. On uplink, MS sends

access burst on assigned PTCCH. Network obtains TA from the burst.

Then network analyses the received TA and provides new TAs for all MSs that perform

TA update on this PDCH. New TAs is sent through downlink signaling message on

PTCCH/D. Network also can send TA in packet TA/power control message and packet

uplink acknowledged/negative message on PACCH.





2-153

Huawei GPRS BSS supports:

Continuous TA update procedure

Quick TA initial value

2.3.11 Measurement Report

Network can request MS to send MR. There are two types of MR:

I. Network control (NC) MR

Network carries relative network control parameters carried in PSI5 broadcast on

PBCCH, SI13 on BCCH, and PSI13 on PACCH to control MS.

Under NC1 or NC2 mode, MS should carry out NC measurement and indicate MR

period in PSI5.

II. Extended MR

Network can order MS to send extended MR. MS sending extended MR is controlled by

XT Measurement Order parameter. This parameter is contained in PSI5 or order packet

measurement message. Network broadcasts PSI5 on PBCCH, or sends order packet

measurement message on PCCCH or PACCH to address a specified MS. The value of

parameter EXT Measurement Order should be EM0, EM1.

When MS is under EM1 mode, it should carry out EM measurement. MR period isspecified in field EXT Reporting Period in PSI5 or in order packet measurement

message. After MS receives order of EM MR, it starts timer T3178 according to the

instructed report period. When T3178 is active, MS can reselect a new cell which is in

EM1 mode. If T3178 timeout duration is greater than the report period indicated in the

new cell, MS should immediately use this period to restart T3178. If timer timeout

duration is less than this period, then T3178 continues working.

Uplink MR is supported. The priority and quality of uplink transmission signal from MS

to site calculated by BTS is sent to PCU through inband signaling of Abis interface

TRAU frame and is used to generate MR.

2.3.12 Supported Flow Control

Gb interface and Um interface have different physical mediums and transfer protocols,

which leads to their different transfer rates. The data transfer rate of Gb interface is

greater than that of Um interface. Moreover, in downlink data transfer, the data transfer

through Um interface is restricted by MS multislot capacity, radio quality, whether there

is radio channel available in the cell. Thus the transfer rate is not constant and it is

necessary to implement flow control for downlink data.





2-154

When cell is in normal state, PCU starts flow control procedure: PCU periodically

reports the cell bucket size and the cell bucket rate according to the state of radio

packet channel in the cell, and reports MS bucket size and bucket rate according radio

resource occupation state of MS.

SGSN adjusts the downlink data rate of this cell/MS according to the reported

parameters, which is the purpose of downlink data flow control.





2-155

Note:

Cell bucket refers to the maximum packet data quantity that allowed being stored. It varies from the

number of packet channels in the cell. MS bucket refers to the maximum packet data quantity that allowed being stored. It varies from the

number of packet channels assigned to MS.

Bucket rate is data transfer rater. Huawei PCU system can implement downlink data flow control,

report the bucket size and bucket rate of the current cell/MS to SGSN, and adjust the reported

parameters according to the changes of cell packet resource and MS resource occupation.

Uplink flow control supports refusal of immediate assignment on CCCH.

For downlink flow control, BVC downlink flow control and MS downlink flow control

are supported.

2.3.13 Supported Dynamic Handover between TCH and PDCH

At the early stage of GPRS service, GSM network is usually updated to support GPRS

service due to the shortage of radio frequencies. In order to reduce the effect on original

GSM circuit switched speech services caused by GPRS service, Huawei GPRS BSS

supports the dynamic handover between TCH and PDCH.

1) Supporting the handover from TCH to PDCH during the establishment of TBF.

Huawei GPRS BSS classifies channel attribute into fixed packet service channel, voice

service channel and dynamic channel. Fixed packet service channel is dedicated for

packet data service, such as PBCCH, PCCCH, and PDCH; voice traffic channel is

dedicated for voice service, such as TCH, BCCH, SDCCH; and the dynamic channel is

voice TCH at its initialization stage. It can be converted between TCH and PDCH.

When there is more packet traffic and the speech channels are relatively idle, PCU will

request the BSC to convert the dynamic channel into the dynamic packet data channel.

Whereas when BSC determines the speech channels are busy, it can also request

PCU to return the converted dynamic channel and use it again as speech channel. In

this process, the speech service is given the priority over the packet service to

guarantee the original speech services.

2) Supporting inter-cell PDCH sharing on the same RPPU

2.3.14 Supported Packet Access Function

When MS upper layers have data to be transferred, MS RLC/MAC will initiate packet

access. MS packet access types are as follows: Short access, phase I access, phase II

access, single block not establishing TBF access, paging response, cell update,

mobility management.





If the data to be transferred is less than 8 RLC blocks, the MS channel request

type is short access. The number of data packets is calculated according to CS-1.

If the data to be transferred is more than 8 RLC blocks and RLC mode is required

to be acknowledged mode, then MS channel request type is phase I access or phase II access.

If MS MR is to be transferred, then the channel request type is monolith not

establishing TBF access.

For channel request type of paging response, cell update, and mobility

management, they are usually processed as phase I or phase II access.

For short access and phase I access, radio resource is assigned for MS in the first time

(such as TFI, dynamic assignment of USF or list of fixed assignment of radio block

position list)

For two-phase access channel request, the first request is for assigning a radio block

for MS. MS sends packet resource request message on assigned sigle radio block for

second resource assignment (including TFI, USF or radio block position list). Then MS

begins to transfer data on assigned resource. The packet channel request is an access

burst with 8 bit or 11 bit, so it carries a little of information. While the packet resource

request is an RLC/MAC signaling packet with CS-1, it can carry relatively more

information (including MS TLLI, MS multislot capacity, radio priority). These kinds of

information are helpful in assigning appropriate resource for MS.

M900/M1800 PCU supports all these access types. For access types such as paging

response, cell update, and mobility management, it processes them by regarding them

as two-phase ones.

Documents

BSS Function Description