LTE 37.901 Throughput

Sandy Fraser

April 2015

Taipei

Page

Agenda – What Factors Determine Throughput

– 37.901 General Conditions

– 37.901 Individual test case examination

– What the UXM can offer for 37.901

– Summary

September 2014 2

Page

LTE throughput

– The UE category

– The Cell Bandwidth, variable from 1.4MHz to 20MHz

– How much space is allocated to the PDCCH (CFI setting)

– The number of Resource Blocks (RB’s) or Resource Block Groups (RBG’s)

– The allocation of Sub-Frames (SF’s), full allocation = 10

– The modulation coding scheme (I-MCS)

– Transport Block Size (I-TBS) – defined block data rates

– The number of spatially multiplexed data streams or codewords, SISO or

MIMO, 2*2, 4*2, 4*4 etc

– Whether Carrier Aggregation is employed

– In the real world, the channel conditions, noise, interference, number of

users all contribute

3

What are the prime factors which determine data rates?

September2014

Page

Real world Conditions affecting LTE/LTE-A device performance Channel and Network factors that impact

mobile device performance

– Fading Conditions

– Doppler Speed

– Degree of Spatial Diversity

– Noise and interference

conditions

– Transmission Mode used

– Influence of re-transmissions

and adaptive modulation and

coding

The CSI/CQI tests measure the UE’s ability to

accurately assess the channel conditions

4

Interference

Noise

Doppler

Multipath

Fading

Adaptive modulation

and coding schemes

September2014

Page

L1 and PHY versus Application Layer TPut

5

Comparison of 37.901 with 36.521 Section 8,9

36.521 S8 37.901

Channel Model Fading, Noise, OCNG Fading, Noise

Test purpose Up to MAC All Layers

Test Specifications 70% of theoretical

maximum allocation

No PASS/FAIL

Allocations FRC Follow CQI,PMI,RI

Data type Fixed MAC Padding UDP, FTP (TCP)

September2014

Page

37.901

6

Summary

– No test limits or specifications – Yet!!

– Tests for FTP/TCP and UDP

– Various fading and noise levels to force re-transmissions, rank

transitions etc

– 37.901 will NOT test fully throughput – no max data rates

– No Carrier Aggregation tests – Yet!!

– 37.901 testing is MUCH simpler with integrated fading, noise and

FTP server

September2014

Page

Agenda – What Factors Determine Throughput

– 37.901 General Conditions

– 37.901 Individual test case examination

– What the UXM can offer for 37.901

– Summary

September 2014 7

Page

37.901

8

– Definition for reference Laptop for tethered tests – no SW which

could hinder throughput

– Some statements on USB and drivers – very obvious.

– Measurement points – for either tethered or stand along UE’s

– Data rates measured without IP headers – so IP address is

important (sec 5.1.2)

– Network restrictions (laptop to be stand-alone, not connected to a

company network

key points v11.11.0 (2014-06)

September2014

Appropriate RF

connection based

upon UE receiver type

including fading and

AWGN

Application

ServerDUT

Appropriate UE to PC interface

Modem or Network Interface

Connection (NIC) and any

associated cabling as

recommeded by the UE

manufacturer for the intended

use by the customer/user

Laptop with Data

Client Test

Application

SS

RF IP

Page

37.901

– Provides application layer UE data throughput performance

measurements for both HSPA and LTE

– Only single Tx, Static Channel Uplink

– Static Channel Bi-directional Uplink and Downlink in Stress

Performance. Either:

• Bi-directional UDP or

• Alternating FTP

– No Carrier Aggregation tests

– No pass/fail verdicts, no test limits

– Defines signal levels, various noise levels, interference, various

fading profiles

9

key points v11.11.0 (2014-06)

September2014

Page

37.901

– UDP because it can be tested bi-directional without the other

direction interfering, and can be used to characterize RTP

• Downlink only, uplink only, bi-directional (concurrent)

– TCP/IP using FTP, because it effectively tests all forms of TCP/IP

underlying protocol such as SFTP and HTTP

• Downlink only, uplink only, bi-directional (concurrent and

alternating)

– SFTP, HTTP use TCP/IP so are not considered separately

– VoIP (RTP-based) has a wide variety of possible test setup’s, but in

testing UDP alone, the underlying RTP (although NOT GBR

aspects) is tested

10

key points v11.11.0 (2014-06)

September2014

Page

E7515A UXM Wireless Test Set

11

Bi-directional TCP End 2 End Carrier Aggregation Throughput

Cat 6 device with

maximum UL and DL

bi-directional TCP

Maximum DL is

290Mbps, Maximum

UL is 45Mbps

45Mbps UL

290Mbps UL

September2014

Page

37.901

12

Table 5.5.4.5-2: Test Points for LTE

Propagation

Condition

Geometry Justification

Static No interference

Note 1

To check that upper-layers do not constrain data throughput

EPA5 20dB To exhibits large TBS variations (see clause 5.5.4.3) and very

common scenarios for high-data rate requiring processing capability

EVA5 10dB EVA occurs frequently in deployments

EVA70 20 dB Adds EVA70, high SNR coverage which is common in low

frequency(<1GHz) band networks

EVA200 20 dB Covers high Doppler, high SNR scenario which is common in high

frequency (≥1GHz)band networks

ETU70 0dB Fast variations and most common high-dispersion case

ETU300 0dB A high BLER scenario may trigger higher layer retransmissions, and

also addresses the high speed scenario in the work item objective

Note 1:In the performance report, the tester shall indicate for the ‘No Interference’ condition, the

following note: In case of 'no interference', the throughput is expected to be maximal. This may

be the maximum theoretical throughput or below. In the latter case it cannot be distinguished,

whether UE limitations, or signal generator limitations with respect to EVM, or both contribute to

this.

September2014

Page

37.901 General test setup

– More common setup parameters listed in backup slides at end of

presentation

– LTE MAC - No Periodic BSR or PHR, No DRX. B.2.2.2

– LTE RLC – Acknowledged Mode only, Setup as per table 4.8.2.1.2.2-1 of TS

36.508. B.2.2.3

– LTE PDCP – No Compression. B.2.2.4

– LTE System Information as per TS36.508 section 4.4.3.4 and 4.4.3.2. B.2.2.5

– PMI is set to follow UE reports. B.2.2.6

– RI is set to follow UE reports. B.2.2.7

– CQI is set to follow the CQI reported by the UE, but the CQI to MCS mapping

can vary depending on Cell BW, and the number of antenna’s used. B.2.2.1

13

Applies to all LTE tests

September2014

Page

CQI allocation tables

CQI

Index

1.4MHz

Single

1.4MHz

Dual

3MHz

Single

3MHz

Dual

5MHz

Single

5MHz

Dual

10MHz

Single

10MHz

Dual

15MHz

Single

15MHz

Dual

20MHz

single

20MHz

Dual

0 DTX DTX DTX DTX DTX DTX DTX DTX DTX DTX DTX DTX

1 0 0 0 0 0 0 0 0 0 0 0 0

2 0 0 0 0 0 0 1 0 0 0 1 0

3 2 1 2 2 2 2 3 2 2 2 3 2

4 4 3 4 4 4 4 5 4 5 4 5 4

5 6 5 6 6 6 6 7 7 7 7 7 7

6 8 7 8 8 8 8 9 9 9 9 9 9

7 11 10 11 11 11 11 12 12 12 12 12 12

8 13 12 13 13 13 13 14 14 14 14 14 14

9 14 14 16 15 16 15 16 16 16 16 16 16

10 17 17 18 18 19 18 19 19 19 19 19 19

11 20 19 21 20 21 20 22 21 22 21 22 21

12 21 21 23 22 23 22 24 23 24 23 24 23

13 23 23 25 24 25 24 27 26 27 26 26 26

14 25 24 27 26 27 26 28 27 28 27 28 27

15 27 25 27 27 27 27 28 27 28 27 28 27

14

Depends on TM, Cell BW and antennae configuration

Amalgamation of 37.901 B.2.2.1 tables

IMCS values relate to only SF1-4,6-9 inclusive

IMCS values for SF0 and 5 must be set to DTX September2014

Page

Scheduling>Subframes Config

Manual CQI to MCS

Mapping allows non-

3GPP CQI index

settings

15 September2014

Page

Agenda – What Factors Determine Throughput

– 37.901 General Conditions

– 37.901 Individual test case examination

– What the UXM can offer for 37.901

– Summary

September 2014 16

Page

37.901

17

Test Cases LTE UDP Downlink

37.901 Title

A.3.3.1 LTE / UDP DL / PDSCH Single Antenna Port Performance (CSRS)

A.3.3.2 LTE / UDP DL / PDSCH Transmit Diversity Performance (CSRS)

A.3.3.3 LTE / UDP DL / PDSCH Open Loop Spatial Multiplexing Performance (CSRS)

A.3.3.4 LTE / UDP DL / PDSCH Closed Loop Spatial Multiplexing Performance (CSRS)

A.3.3.5 LTE / UDP DL / PDSCH Single-layer Spatial Multiplexing Performance (Port 5, UE-Specific RS)

A.3.3.6 LTE / UDP DL / PDSCH Single-layer Spatial Multiplexing Performance (Port 7 or 8, UE-Specific RS)

A.3.3.7 LTE / UDP DL / PDSCH Dual-layer Spatial Multiplexing Performance (port 7 and 8, User-Specific RS)

September2014

Page

37.901

18

Test Cases LTE FTP Downlink

37.901 Title

A.3.2.1 LTE / FTP DL / PDSCH Single Antenna Port Performance (CSRS)

A.3.2.2 LTE / FTP DL / PDSCH Transmit Diversity Performance (CSRS)

A.3.2.3 LTE / FTP DL / PDSCH Open Loop Spatial Multiplexing Performance (CSRS)

A.3.2.4 LTE / FTP DL / PDSCH Closed Loop Spatial Multiplexing Performance (CSRS)

A.3.2.5 LTE / FTP DL / PDSCH Single-layer Spatial Multiplexing Performance (Port 5, UE-Specific RS)

A.3.2.6 LTE / FTP DL / PDSCH Single-layer Spatial Multiplexing Performance (Port 7 or 8, UE-Specific RS)

A.3.2.7 LTE / FTP DL / PDSCH Dual-layer Spatial Multiplexing Performance (port 7 and 8, User-Specific RS)

September2014

Page

A.3.2.4 FTP DL Closed Loop MIMO

19

Parameter Unit All Tests

Downlink power

allocation

dB -3

dB -3 (Note 1)

at antenna port dBm/15kHz -85 (Note 2)

at antenna port dBm/15kHz -98

Transmission mode 4

Reporting periodicity ms Npd = 5

cqi-pmi-ConfigurationIndex 4

ri-ConfigurationInd 1 (Note 4)

CQI delay ms 8 for FDD

10 for TDD (Note 3)

Reporting mode PUCCH 1-1

CodeBookSubsetRestriction bitmap 111111

Note 1:

Note 2: is applied to only Test Number 1 in Table A.3.2.4.3-2: Test points for Closed Loop

SpatialMultiplexing Downlink Testing

Note 3: If the UE reports in an available uplink reporting instance at subframe SF#n based on PMI

and CQI estimation at a downlink subframe not later than SF#(n-4), this reported PMI and

wideband CQI cannot be applied at the eNB downlink before SF#(n+4).

Note 4: To avoid the ambiguity of SS behaviour when applying CQI and PMI during rank

switching, RI reports are to be applied at the SS with one subframe delay in addition to

Note 3 to align with CQI and PMI reports.

A

B

ocN

1BP

sE

Table A.3.2.4.3-1: Test Parameters for Closed Loop

Spatial Multiplexing Downlink Testing

sE

September2014

Page

A.3.2.4 FTP DL Closed Loop MIMO

20

Initial conditions

Channel bandwidth Note1

Test parameters for each bandwidth

Test

Number

Reference test

point

Note 2

Propagation Conditions

Note 2

SNR (dB)

Note 2

Correlation

1 LTE-1 Static No

interference

Note 3

N/A

2 LTE-2 EPA5 20 Low

3 LTE-3 EVA5 10 Low

4 LTE-4 ETU70 0 Low

5 LTE-5 ETU300 0 Low

Note 1: See Annex B.3 for the recommended channel bandwidth

Note 2: The test points is according to Table B.1.2-1in Annex B.1.2.

Note 3: In the performance report, the tester shall indicate for the ‘No Interference’ condition,

the following note: In case of 'no interference', the throughput is expected to be

maximal. This may be the maximum theoretical throughput or below. In the latter

case it cannot be distinguished, whether UE limitations, or signal generator

limitations with respect to EVM, or both contribute to this.

Table A.3.2.4.3-2: Test points for Closed Loop Spatial

Multiplexing Downlink Testing

September2014

Page

A.3.2.4 FTP DL Closed Loop MIMO

• Setup the SS/OBT to respond to the UE’s reported CQI, RI and PMI

• Setup the SS/OBT for the correct BW, RMC, Fading, Noise etc

• Using the FTP client, begin FTP download from the application server of a file sufficient in

size for the test duration (60 seconds for static, 36.521 Annex G3.5 for Faded tests – 14 to

1500seconds), see 37.901 Table A.3.1-1. The faded test times vary depending on: Fading

profile, Antenna configuration, Correlation, RMC used etc.

• Note: HSPA tests use 60 seconds for static and 164s for all faded tests

• Record Throughput T result for this first iteration

• Repeat for a further 2-5 iterations

• Calculate and record the Average of the 5 iterations

• Count and record the overall number of ACK and NACK/DTX on the PUSCH/PUCCH during

the test interval.

• Repeat steps above for each subtest in Table A.3.2.4.3-2 on the previous page

• There are no pass/fail limits

21

Test process

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

22

Generally flat at MAC, with

little TCP variations

No MAC re-transmissions to

give MAX Tput of 50.73Mbps

Step 1 Static No Noise, Throughput/BLER 10MHz

September2014

Page

Why 50.73Mbps and not 73.39Mbps?

August 2014 23

Effects of no SF0, 5, follow CQI and lower I-TBS

Full DL Throughput

BW 1.4 3 5 10 15 20

RBs 6 15 25 50 75 100

MAX I_MCS 28 28 28 28 28 28

MAX I_TBS 26 26 26 26 26 26

Sub-Frames 10 10 10 10 10 10

I_TBS table bits 4392 11064 18336 36696 55056 75376

Tput 1*CW 4392 11064 18336 36696 55056 75376

TPut 2*CW 8784 22128 36672 73392 110112 150752

37.901 Throughput

Max Follow I-MCS 25 27 27 27 27 27

Max Follow I_TBS 23 25 25 25 25 25

Sub-Frames 8 8 8 8 8 8

I-TBS table bits 3496 9528 15840 31704 46888 63776

Tput 1*CW 2796.8 7622.4 12672 25363.2 37510.4 51020.8

TPut 2*CW 5593.6 15244.8 25344 50726.4 75020.8 102041.6

Page

Actual 37.901 results with UXM – A.3.2.4

24

Rank 2, CQI 15, perfect

channel, no noise, maximum

throughput

No MAC re-transmissions

Step 1 Static No Noise, CSI 10MHz

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

25

Highly variable

throughput and BLER,

uses all 4 transmissions

The only test step in

A.3.2.4 which really

needs averaging

Step 2 EPA5 20dB SNR, Throughput/BLER 10MHz

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

26

Widely ranging

CQI, with only 1

or 2 transitions

into Rank =1

Step 2 EPA5 20dB SNR, CSI 10MHz

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

27

Narrower variation

range due to more

noise, using all 4

transmissions

Step 3 EVA5 10dB SNR, Throughput/BLER 10MHz

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

28

Oscillation between rank 1and 2,

two distinct groups of CQI

(10-13 Rank 1, 5-13 Rank 2)

Step 3 EVA5 10dB SNR, CSI 10MHz

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

29

Tighter grouping for

BLER and CSI due to

high noise and no

Rank Transitions

Step 4 ETU70 0dB SNR, Throughput/BLER 10MHz

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

30

Step 4 ETU70 0dB SNR, CSI 10MHz

Tighter grouping for

BLER and CSI due to

high noise and almost

no Rank Transitions

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

31

Only slight drop in

Tput over ETU70

with same noise

Step 4 ETU300 0dB SNR, Throughput/BLER 10MHz

September2014

Page

Actual 37.901 results with UXM – A.3.2.4

32

Step 4 ETU300 0dB SNR, CSI 10MHz

Tight grouping for

BLER and CSI due to

high noise and no

Rank Transitions

September2014

Page

37.901

33

Test Cases LTE UL, Stress, Power Sweep

37.901 Title

A.3.4 LTE / FTP UL / PDSCH Single Antenna Port Performance (CSRS)

A.3.5 LTE / UDP UL / PDSCH Transmit Diversity Performance (CSRS)

A.3.6.1 LTE / Stress Test Performance / PDSCH Transmit Diversity Performance (CSRS)

A.3.6.2 LTE / Stress Test Performance / PDSCH Open Loop Spatial Multiplexing Performance (CSRS)

A.3.7.1 LTE / UDP Power Sweep Performance / PDSCH Transmit Diversity Performance (CSRS)

A.3.7.2 LTE / UDP Power Sweep Performance / PDSCH Open Loop Spatial Multiplexing Performance (CSRS)

• The UL tests are very simple static channel

• Stress Tests are two single stage tests with either bi-directional UDP or

alternating direction FTP

• Power Sweep tests utilize a number of tests (N) with varying Ior values. In the

A.3.7.2 case, the Ior value decrements down in 2dB steps from -78dB to

REFSENS +6dB, all using EVA70. Ior Compensation is made for the Cell BW

September2014

Page

A.3.7.2 UDP Open Loop Power Sweep

34

Test Conditions

Parameter Unit All Tests

Downlink

power

allocation

dB -3

dB -3 (Note 1)

Transmission mode 3

Reporting interval ms 5

CQI delay ms 8

Reporting mode PUCCH 1-0

Note 1:

A

B

1BP

Table A.3.7.2.3-1: Test Parameters for Open Loop Spatial Multiplexing UDP Power Sweep Testing

Channel bandwidth

BWChannel [MHz] 1.4 3 5 10 15 20

Ior power level offset

(dB) -9.2 -5.2 -3.0 0.0 1.8 3.0

Table A.3.7.2.3-3: Ior Power level offsets for Open Loop Spatial Multiplexing UDP Power Sweep Testing

September2014

Page

A.3.7.2 UDP Open Loop Power Sweep

35

Test Conditions

Initial conditions

Channel bandwidth Note1

Test parameters for each bandwidth

Test

Number

Propagation

Conditions

Ior (dBm) Correlation

1 EVA70 -60 Low

2 EVA70 -62 Low

3 EVA70 -64 Low

4 EVA70 -66 Low

5 EVA70 -68 Low

6 EVA70 -70 Low

7 EVA70 -72 Low

8 EVA70 -74 Low

9 EVA70 -76 Low

10 EVA70 -78 Low

11 EVA70 -80 Low

12 EVA70 -82 Low

Table A.3.7.2.3-2: Test Points for Open Loop

Spatial Multiplexing UDP Power Sweep Testing

Note 1: See Annex B.3 for the recommended

channel bandwidth

Note 2: Determine if throughput increases as

the signal level is increased in relation

to the UE noise floor and remains

consistent across Ior values within a

reasonable tolerance once the

throughput has reached a maximum.

Note 3: Ior (dBm) power levels are specified for

10 MHz channel bandwidth. For other

channel bandwidths, add the offset

defined in Table A.3.7.2.3-3.

Note 4: In the performance report, the tester

shall indicate that the throughput is

expected to reach a maximum. This

may be the maximum theoretical

throughput or below. In the latter case it

cannot be distinguished, whether UE

limitations, or signal generator

limitations with respect to EVM, or both

contribute to this.

September2014

Page

A.3.7.2 UDP Open Loop Power Sweep

36

Test Process

• Setup the SS/OBT to respond to the UE’s reported CQI, RI and PMI

• Setup the SS/OBT for the correct BW, RMC, Fading profile etc

• Using the UDP client, begin UDP download for the test duration (60 seconds for static,

36.521 Annex G3.5 for Faded tests), see 37.901 Table A.3.1-1. The faded test times vary

depending on:

• Fading profile

• Antenna configuration

• Correlation

• RMC used etc

• Record Throughput T result.

• Count and record the overall number of ACK and NACK/DTX on the PUSCH/PUCCH during

the test interval.

• Repeat steps above for each subtest in Table A.3.7.2.3-2 on the previous page

• Determine if throughput increases as the signal level is increased in relation to the UE noise

floor and remains consistent across Ior values within a reasonable tolerance once the

throughput has reached a maximum.

• There are no pass/fail limits

September2014

Page

37.901

37

Test Cases LTE DL versus SNR

37.901 Title

A.3.8.1 LTE / UDP DL vs SNR PDSCH Transmit Diversity Performance (CSRS))

A.3.8.2 LTE / UDP DL vs SNR PDSCH Open Loop Spatial Multiplexing Performance (CSRS)

A.3.8.3 LTE / UDP DL vs SNR PDSCH Closed Loop Spatial Multiplexing Performance (CSRS)

A.3.8.4 LTE / UDP DL vs SNR PDSCH Single Layer Closed Loop Spatial Multiplexing Performance (CSRS)

• Each of the 4 tests above consists of either 18 or 21 separate test steps

• The test steps are in three groups using Static Channel, EPA5 or EVA70.

• Each group has several levels of Ior and SNR

September2014

Page

A.3.8.3 UDP DL vs SNR

38

Test Conditions Parameter Unit All Tests

Downlink power

allocation

dB -3

dB -3 (Note 1)

at antenna port dBm/15kHz As needed per Test No

Transmission mode 4

Reporting interval ms 5 (Note 2)

CQI delay ms 8 for FDD

10 for TDD (Note 3)

Reporting mode PUSCH 1-2

CodeBookSubsetRestriction

bitmap

111111

Note 1:

Note 2: SS schedules the UL transmission to carry the PUSCH CQI feedback via PDCCH DCI

format 0 with CQI request bit set to 1

Note 3: If the UE reports in an available uplink reporting instance at subframe SF#n based on

PMI and CQI estimation at a downlink subframe not later than SF#(n-4), this reported

PMI and wideband CQI cannot be applied at the eNB downlink before SF#(n+4).

A

B

ocN

1BP

Table A.3.8.3.3-1: Test Parameters for Closed Loop Spatial

Multiplexing Downlink Testing September2014

Page

A.3.8.3 UDP DL vs SNR

39

Test Conditions Initial conditions

Channel

bandwidth

Note1

Test parameters for each bandwidth

Test No Propagation

Conditions

Ior

(dBm)

SNR (dB) Correlation

1

Note 3

Static -50 No

Interference

Note 2

N/A

2 Static -55 25 N/A

3 Static -60 20 N/A

4 Static -65 15 N/A

5 Static -70 10 N/A

6 Static -75 5 N/A

7 Static -80 0 N/A

8 EPA5 -50 No External

Noise

Low

9 EPA5 -55 25 Low

10 EPA5 -60 20 Low

11 EPA5 -65 15 Low

12 EPA5 -70 10 Low

13 EPA5 -75 5 Low

14 EPA5 -80 0 Low Table A.3.8.3.3-2: Test points for Closed Loop

Spatial Multiplexing Testing

15 EVA70 -50 No

External

Noise

Low

16 EVA70 -55 25 Low

17 EVA70 -60 20 Low

18 EVA70 -65 15 Low

19 EVA70 -70 10 Low

20 EVA70 -75 5 Low

21 EVA70 -80 0 Low

Note 1: See Annex B.3 for the recommended channel

bandwidth.

Note 2: In the performance report, the tester shall indicate

for the ‘No Interference’ condition, the following

note: In case of ‘No Interference’, the throughput

is expected to be maximal. This may be the

maximum theoretical throughput or below. In the

latter case it cannot be distinguished, whether UE

limitations, or signal generator limitations with

respect to EVM, or both contribute to this.

Note 3: If this test point is part of another test there is no

need to repeat. The previous result can be used.

September2014

Page

A.3.8.3 UDP DL vs SNR

40

Test Process

• Setup the SS/OBT to respond to the UE’s reported CQI, RI and PMI

• Setup the SS/OBT for the correct BW, RMC, Fading, Noise etc, connection

diagram 36.508 Annex A, Figure A.10 (see backup slides)

• Using the UDP client, begin UDP download for the test duration (60

seconds for static, 36.521 Annex G3.5 for Faded tests), see 37.901 Table

A.3.1-1. The faded test times vary depending on:

• Fading profile

• Antenna configuration

• Correlation

• RMC used etc

• Record Throughput T result.

• Count and record the overall number of ACK and NACK/DTX on the

PUSCH/PUCCH during the test interval.

• Repeat steps above for each subtest in Table A.3.8.3.3-2 on the previous

page

• There are no pass/fail limits

September2014

Page

Agenda – What Factors Determine Throughput

– 37.901 General Conditions

– 37.901 Individual test case examination

– What the UXM can offer for 37.901

– Summary

September 2014 41

Page

E7515A UXM Wireless Test Set Carrier Aggregation end-to-end traffic

PDCP

RLC

Lower

MAC

Lower

MAC

MAC

PHY/

RF

PHY/

RF

IP

Uplink

50 Mbps

Downlink

300 Mbps

DL CC1

150

Mbps

UL

50

Mbps

L1

L2

IP

DL CC2

150

Mbps

September2014

Page

UXM CSI RRC Config Setup

43

Enable either Periodic or

Aperiodic CSI reporting with

necessary parameter and

reporting mode settings

September2014

Page

UXM – Flexible Allocations Static, Adaptive, Random PMI/RI

44

Static, Adaptive

and Random

PMI/RI allocations

Sub-frame by sub-

frame allocations

based on UE CQI

report values

Easy to eliminate

SF0 and SF5 for

testing section 8

September2014

Page

UXM Fading and Noise

AWGN

3GPP standard

Correlation Modes

and Fading Profiles

including CQI for

36.521 section 9

OCNG

45 September2014

Page

Extensive throughput testing

Easy access to

Impairments and

power control

46 September2014

Page

CSI reporting

Easy access to

Impairments and

power control

during CQI testing

PMI Codebook

selection – this

example 4*2 with

some codebook

indices deselected

47 September2014

Page

CSI reporting with Carrier Aggregation

48

Two completely

independent cells,

in this case (CA)

PCC and SCC

Results for each

Component

Carrier

Statistical results

for each codeword

September2014

Page

Agenda – What Factors Determine Throughput

– 37.901 General Conditions

– 37.901 Individual test case examination

– What the UXM can offer for 37.901

– Summary

September 2014 49

Page

L1 and PHY versus Application Layer TPut

50

Comparison of 37.901 with 36.521 Section 8,9

36.521 S8 37.901

Channel Model Fading, Noise, OCNG Fading, Noise

Test purpose Up to MAC All Layers

Test Specifications 70% of theoretical

maximum allocation

No PASS/FAIL

Allocations FRC Follow CQI,PMI,RI

Data type Fixed MAC Padding UDP, FTP (TCP)

September2014

Page

37.901

51

Summary

– No test limits or specifications – Yet!!

– Tests for FTP/TCP and UDP

– Various fading and noise levels to force re-transmissions, rank

transitions etc

– 37.901 will NOT test fully throughput – no max data rates

– No Carrier Aggregation tests – Yet!!

– 37.901 testing is MUCH simpler with integrated fading, noise and

FTP server

September2014

Page

End Main Presentation

52 September2014

Page

What next for Throughput testing?

53 September2014

Page

New UE Categories

54

Table 4.1-1 from 3GPP 36.306 up to Release 10

September2014

UE Category Maximum number of DL-SCH

transport block bits received

within a TTI (Note 1)

Maximum number of bits of a

DL-SCH transport block

received within a TTI

Total number of soft

channel bits

Maximum number of

supported layers for

spatial multiplexing in DL

Category 1 10296 10296 250368 1

Category 2 51024 51024 1237248 2

Category 3 102048 75376 1237248 2

Category 4 150752 75376 1827072 2

Category 5 299552 149776 3667200 4

Category 6 301504 149776 (4 layers)

75376 (2 layers)

3654144 2 or 4

Category 7 301504 149776 (4 layers)

75376 (2 layers)

3654144 2 or 4

Category 8 2998560 299856 35982720 8

Commercially

available UE’s NOW!

Page

New UE Categories

55

Table 4.1-1 from 3GPP 36.306 beyond Release 10

September2014

UE Category Maximum number of DL-SCH

transport block bits received

within a TTI (Note 1)

Maximum number of bits of a

DL-SCH transport block

received within a TTI

Total number of soft

channel bits

Maximum number of

supported layers for spatial

multiplexing in DL

Category 0 1000 1000 25344 1

Category 9 452256 149776 (4 layers)

75376 (2 layers)

5481216 2 or 4

Category 10 452256 149776 (4 layers)

75376 (2 layers)

5481216 2 or 4

Category 11 603008 149776 (4 layers, 64QAM)

195816 (4 layers, 256QAM)

75376 (2 layers, 64QAM)

97896 (2 layers, 256QAM)

7308288 2 or 4

Category 12 603008 149776 (4 layers, 64QAM)

195816 (4 layers, 256QAM)

75376 (2 layers, 64QAM)

97896 (2 layers, 256QAM)

7308288 2 or 4

Category 13 391632 195816 (4 layers)

97896 (2 layers)

3654144 2 or 4

Category 14 391632 195816 (4 layers)

97896 (2 layers)

3654144 2 or 4

Category 15 3916560 391656 47431680 8

R&D available UE’s

NOW!

Commercially

available UE’s NOW!

Page

CAT 0 support for IoT

E7515A UXM

Keysight

Confidential 56

Low Power, low cost, high volume …

Release 8 CAT1 Release 12 Cat 0 Release 13

DL peak data rate 10 Mbps 1 Mbps ~200 kbps

UL peak data rate 5 Mbps 1 Mbps ~200 kbps

Max No. of spatial layers 1 1 1

No. of UE Rx chains 2 1 1

Duplex mode Full duplex Half duplex (optional) Half duplex (optional)

UE Rx bandwidth 20MHz 20MHz 1.4MHz

Max UE Tx power 23 dBm 23 dBm ~20dBm

• Low Power - Extended paging cycle and transaction cycle (not yet defined)

• Low cost - Reduced bandwidth, reduced support for DL TMs, spec relaxations (MCS support, TBS size

etc.)

• High volume - IPv6 addressing (not MSISDN), eUICC

Page

New UE Categories

57

UXM plot of Cat 9, 3CC device, IP data

September2014

Page

New UE Categories

58

UXM plot of Cat 9, 3CC device, IP data

September2014

20 MHz Bandwidth

LTE Carrier #3

150Mbps IP data

LTE-A Carrier Aggregation Solution: 10 or 20 MHz fragments aggregated to get 30 - 60MHz channel bandwidth

20 MHz Bandwidth

20 MHz Bandwidth

Aggregated

Carriers LTE Carrier #2

LTE Carrier #1

60 MHz Bandwidth

450 Mbps IP data 150Mbps IP data

150Mbps IP data

Page

(F)eICIC

– Test throughput behavior in networks

employing both large and small cells

– Perform 3GPP demod performance,

UE meas, and functional performance tests,

or use custom settings to explore beyond the specs

– Designate each UXM cell as Aggressor or Victim

– Specify 36.521-1 or custom ABS patterns

– Verify FeICIC reduced power

and CRS cancellation

– Analyze data throughput

and UE meas results for

ABS, non-ABS or

combined subframes

E7515A UXM

Keysight Restricted 59

Go deeper in functional test

Page

The importance of WiFi + Mixed Aggregation

E7515A UXM

Keysight

Confidential 60

More bandwidth, more aggregation, mixed aggregation

• WiFi Offload – transfer of data from LTE to WiFi

• LTE-U in 3GPP Rel 13 is referred to as Licensed Assisted Access (LAA)

- LTE Use of 2.4 and 5GHz unlicensed spectrum, eg Bands 125, 127

- Interoperability with WiFi neighbour cells

• Inter-RAT carrier aggregation, eg LTE + WiFi

• Mixed FDD-TDD carrier aggregation

LTE WiFi

LTE licensed

WiFi, unlicensed

• LTE + WiFi aggregation

• Licensed through LTE

• LTE network control of WiFi

• Simultaneous LTE + WiFi = Wider

overall data pipe

Page

What next for Throughput testing? End

61 September2014

Page

Thank you for listening !

Any Questions ?

Page 62