Analysis of Changes in PMSE Spectrum and Transmission ... fileAnalysis of Changes in PMSE Spectrum and Transmission Technology -Our Prediction for PMSE, Operated Under Changed Conditions

Analysis of Changes in PMSE Spectrum and Transmission Technology - Our Prediction for PMSE, Operated Under Changed Conditions - Update 2017WM-01 Current and Future Use of Spectrum by PMSE

4th PMSE Workshop at EuMWProf. Dr.-Ing. Georg FischerInstitute of Electronics Engineering

The birthplace of

1. What’s hot?

2. Review of Digital transmission technology

3. Crest factor and Pulsing of interferers

4. Airband use?

5. How to prepare for the future?

6. Conclusions

2

Content

EUMW2017, WM01, Workshop on PMSE, Nuremberg, 9th Oct. 2017

1. What’s hot?

3

4EUMW2017, WM01, Workshop on PMSE, Nuremberg, 9th Oct. 2017

What’s hot?Interpretation

Transmission Technology• Digital PMSE has arrived in daily use• Practical experience tells that Digitization of PMSE does not allow for less spectrum• Digitization of PMSE is no relief to spectrum shortage by Digital Dividend I+II• Transition to Digital PMSE brings more quality but also new challenges

New Frequency Bands• Duplex gaps of cellular systems get more used, however crackling noise• Airband is under discussion, however compatibility unclear

Convergence in Media transmission• Broadcast and cellular are in fruitful talks• Broadcast has brought in requirements catalogue into 3GPP standardization for 5G• 5G MEDIA initiative founded in Munich 2017• Research: Can PMSE also join the Media convergence? See PMSE-xG project

New Physical layer for PMSE?• Can the 5G physical layer be used for PMSE? Research ongoing…• How does a wideband air interface behave relative to todays narrowband interface?• Some parties are advocating a wideband air interface for PMSE…

2. Review of Digital transmission technology

5

Availability• Drop-outs not acceptable, 100% reliable (telecoms 95%) • Historic events cannot be repeated “Daß Ihre Ausreise…(Genscher)”• No Clicks acceptable, especially in quiet periods (e.g. Digital out of sync…)

Quality• High quality must be captured in production archive as otherwise it is not possible to

derive the various quality levels in distribution• Large compression is thus not possible on production, only on distribution

Latency• The “drummer problem”, roundtrip from wireless mic to wireless IEM should be < 5ms

Telephone 200ms, Cellular 20ms, Satcom 500ms• Communication theory wise: Information source and information sink are collocated!

Battery operation• PMSE equipment (mics, IEM) at artist operated from battery• Part of costumes

6

Review of Digital transmission technologyPMSE specifics


PMSE has special demandsExpectations regarding spectrum savings from digitization of PMSE too high

Findings• Wireless Technology on Production 2x (MIC+IEM)• Wireless Technology on Distribution• Large variety of distribution CODECs• Life productions – it must work reliable!

7

Review of Digital transmission technologyPMSE specifics

Production Distribution

High Quality AudioArchive

(Former Master Tape)

Broadcast

CD/DVD

InternetMP3 downloadMusic streaming

Roundtrip Latency?


IEM

PMSEwireless

wireless

wireless

wireless

What’s not captured in the archive,cannot be distributed later!

8

Review of Digital transmission technologyInformation theoretical view


Model• Information from information source (microphone) is sent to information sink

(mixing console, archive, distribution, IEM)• Goal is to minimize transinformation, it defines spectrum need• Redundancy: we can strip this off without loss of quality• Irrelevance: debatable …who decides? Big brother? The receiver should...• The more irrelevance is identified, the less transinformation, the less spectrum

is needed…

Information source

Information sink

IrrelevanceRedundancy

Transinformation

Error

Spectrum need defined by transinformation

CompressionSource coding

9

Review of Digital transmission technologyDigital wireless transmission system


Information source

Source codingCompression

Channel CodingProtection

Channel Sounding Training sequence

Synchronisation information

Multiplexer

Transmitter

Overhead

ProtocolSignalling

Digital Transmission comes at the cost of overhead

10

Review of Digital transmission technologyPrice to pay for digital – overhead with digital


Digital transmission• Source information is sourced coded (compressed)• Channel coding needed for robustness of transmission• Overhead not negligible!

Raw data

source coded data

Channel codeddata

Coded data includingoverhead

Digital only pays off if we gain more by compression,than we pay by overhead

11

Review of Digital transmission technologyPrice to pay for digital – overhead with digital


Today’s digital PMSE does not provide Studio quality

0,0

0,5

1,0

1,5

2,0

2,5

3,0

3,5

4,0

4,5

5,0

0,10,30,51,02,04,08,016,032,064,0

Mbi

t/s

Compression factor

Codec quality & datarate

CD (reference)

Studio I24bit@96 kSa/s

Studio II24bit@192 kSa/s

PMSE(standard)PMSE

(robust)DABDAB+DRM


Review of Digital transmission technologyTransmission robustness (Channel coding)

Diversity• Makes transmission robust, various flavours

Antenna diversity• Spatial diversity - multiple antenna locations• Polarization diversity - multiple antenna

polarisations at one location• Pattern diversity - different antenna types

Temporal diversity• Channel coding, block codes, interleaving• 2G GSM spreading over 40 ms (20 ms blocks), 3G UMTS 10 ms blocks, 4G LTE

ACK/NACK protocol 4..8 ms• PMSE demands 1..2 ms one way!

Frequency diversity• Frequency hopping in narrowband systems, e.g. GSM across 35…75 MHz• Frequency hopping also provides interference diversity• Wideband interfaces, e.g. W-CDMA/UMTS or OFDM/LTE, width 1.25….20 MHz• Coherence bandwidth of channel typ. 1.5 MHz • PMSE narrowband, only 200 kHz channel, historic reasons, similar analogue FM Radio

time

frequency antenna(space, polarization,

pattern)


Review of Digital transmission technologyDiversity schemes

Antenna diversity• Receiver diversity already in place with PMSE receivers• Could migrate to 2x2 MIMO ?• Complexity impact to Wireless mic?• Massive MIMO? Even more antennas, partly done with distributed antenna systems

Temporal diversity• Not a realistic option for PMSE due to stringent latency requirement• Cellular much less latency critical, so they can use this for robustness

Frequency diversity• Today’s PMSE is very narrowband • Wideband techniques would also provide interference diversity• Cognitive spectrum management already leads to frequency handovers due to

interference

Summary• Only antenna and frequency/interference diversity are realistic options as they do not

imply large latency• Temporal diversity (interleaving) no option due to latency hit!

3. Crest factor and Pulsing of interferers

14

15

Crest factor and Pulsing of interferersProblem description


Characteristics of interferers to PMSE• Often use advanced air interfaces, higher order modulation format• strong fluctuation of momentary power, no constant interferer power• Have Crest, 12..15 dB peak to average

Interference mechanism• If channel sounding training sequence or sync symbols are hurt, then also data is hurt• Chanel sounding and syn symbols are central points of failure• No similar mechanism on analogue• Loss of sync leads to long drop out - 500 ms?

Consequence• Stay away from loss of sync• Apply sufficient margin to noise and interference• Practically digital does not allow for more dense packing of frequencies than analogue

Resync on digital implies long drop out

16

Crest factor and Pulsing of interferersAnalogue versus digital


Digital is better here

Source: Sennheiser

Digital is much worse here

The digital cliff

The analogue cliff

Digital cliff -versus- analogue soft degradation

17

Crest factor and Pulsing of interferersInterference into PMSE


Characteristics of PMSE• PMSE as victim is very vulnerable by interference• Due to high audio quality requirements in PMSE, it is operating at a fairly high SNR

around 20…40 dB, other systems typically operate at low SNR e.g. 7 dB• Due to stringent latency requirements in PMSE (5ms round trip), channel coding by

interleaving can only by spread across a block of 1 ms, thereby not delivering strong temporal diversity gains

• Interference to PMSE through strong peaks will make block decoding fail

Regulation• So far protection levels given as average power levels (interference temperature)• Better specify peak levels

The strong latency and quality requirementsmake PMSE very vulnerable to peaky interference

18

Crest factor and Pulsing of interferersInterference into PMSE


Interference from LTE shoulder• PMSE might be operated next to LTE basestations and terminals, Duplex gap• Side emissions are specified as average power limits, however peak levels are

crucial for PMSE• If quantifying interference peak levels should be assessed

Crest factor at channel 7…10 dB

Crest factor at shoulder 40 dB

Average power on shoulder not meaningful!

LTE BS

4. Airband use?

19

Airband use?Interference situation

New interference situation at L-Band• Multiple systems present in L-Band• DME Distance Measurement Equipment (civil and military versions)• LDACS1 digital avionic radio coming up• Military Radio Systems: JTIDS/MIDS

DME• Double Pulses (pulse pairs), ideally Gaussian shaped• Realistically sharper pulses are used, occupying larger bandwidth• Repetition rates: Ground station 2700 ppps (pulse pairs per second), Military version

TACAN 3600 ppps, Aircraft 150 ppps in the search mode and 30 ppps in the track mode

12 μs

EUMW2017, WM01, Workshop on PMSE, Nuremberg, 9th Oct. 2017 20

Source: Hirschbeck Martin, Interference Mitigation and Channel Estimation for Digital Aeronautical Communication Systems based on OFDM, FAU, July 2016

21

Airband use?Interferers


LDACS• L-Band Digital Aeronautical Communication System• OFDM System• Interleaved with DME

Notations• UAT=Universal Access Transceiver• SSR=Secondary Radar System• JITDS=Joint Tactical Information Distribution System• MIDS=Multi-functional Information Distribution System

Source: Hirschbeck Martin, Interference Mitigation and Channel Estimation for Digital Aeronautical Communication Systems based on OFDM, Dissertation to be defended, FAU, July 2016

22

Airband use?Spectrum assignments in L-Band


Coordination• Forward link FL and reverse link RL coordinated between LDACS and DME


23

Airband use?What about LDACS interference by DME?


LDACS being interfered by DME• Time domain view

• On average every third OFDM Symbol is hurt• Block length in OFDM (e.g. 20 ms?) much higher than in PMSE (2ms)• Larger block length gives higher interleaving gain• Larger block length possible in LDACS as less latency critical


OFDM Symbol120 μs

24

Airband use?Time domain view


DME Interference• DME consist of double pulses with 12us distance• Pulses are nearly Gaussian• DME runs at 2700…3600 ppps (pulse pairs per second)

PMSE as victim• Goal 5 ms roundtrip, 2 ms one way plus 1ms for mixing console• Then in 2 ms there are 7 double pulses• A double pulse occupies roughly 20 us duration of interference• A duration of 20 us * 7= 140 us out of 2 ms would be interfered• 7% of PMSE transmission interfered• If DME pulses hit pilot symbols in PMSE transmission, whole frame would be

immediately lost• In case of incoherent PMSE transmission there would be no pilots, e.g. CPM

incoherent, however incoherent CPM is limited to about 3 bit/s/Hz

Conclusion• DME pulses happen several times in a PMSE frame

The assumption of seldom hits is not true!

25

Airband use?Power levels


Transmit power• TACAN (military version) ground max 4 kW=+66 dBm• DME ground 1 kW• DME Board 100W…500W (2kW militärisch)• PMSE max 50mW, typ 10 mW =+10 dBm• Sensitivity PMSE typ -95 dBm

Relations• DME transmission power factor 1000…40,000 equal 56 dB higher than that of PMSE

Interpretation• PMSE is a system with short frame length and short interleaving• Temporal diversity through interleaving is rarely in place• For quantifying DME interference into PMSE receiver, peak and not average receive

level of interference should be used.

Note• The Shannon bound is valid only for long (infinite) time interleaved systems• PMSE as a latency critical system cannot be described by Shannon and is vulnerable

through peak interference• PMSE cannot profit from interleaving as a protection mechanism

26

Airband use?SNR hit by latency constraint


Theoretical background• Required SNR predicted by Shannon bound assumes infinitely long interleaving• No analytic formula to derive SNR hit by latency constraint• Nevertheless theoretical bounds on hit can be identified• Hit on the order of extra 5 dB to meet latency constraint, under certain cases may go

up to 10 dB

Source: Rachinger Ch., Huber J., Müller R., Comparison of Convolutional and Block Codes for Low Structural Delay, IEEE Transactions on Communications,, Volume: 63, Issue: 12, Pages: 4629 - 4638, 2015

SNR hit5 dB

5. How to prepare for the future?(media convergence…)

27

28

How to prepare for the future?PMSE and broadcast


• Production companies• Need for PMSE

Content& Event

production

Content management

Content distribution

• Broadcaster’s core business!

• Public address• TV/Radio• Internet• Mediathek• CD/DVD

Situation has changed• Broadcasters no longer are present throughout the whole business chain• Broadcasters focus on core business - content management• Broadcasters no longer represent PMSE and Broadcast spectrum inside one institution

• Downloads• Internet• Satellite• Cable• Shop...

The Classical Model for PMSE coexistence with Broadcastno longer works

29

How to prepare for the future?Broadcast via eMBMS inside 5G


Source: BFS IMB5

Broadcast over cellular• New business opportunities• TV Distribution• Software Update• Local advertisements• Fleet Management• Car info• Tourist/Travel info

Research• Project in Germany BFS IMB5• Partners: IRT, NOKIA, FhG IIS, R&S, BR, BMW

Technical approach• Longer CP as in DVB-T/T2• 16.6 / 33.3 us, equal to 5 / 10 km• SFN (Macro Diversity gain)

30

How to prepare for the future?Broadcast via eMBMS inside 5G


Source: BFS IMB5

2030…

DVB-T2 Introduction

2020201920182017201620152014

5G-Standardisation in 3GPP

Pre 5G introducedat Olympic Winter Games in southKorea

5G at OlympicSummer Games in Japan

Follow up projectBFS IMB5

Introduction of convergent

system

Schedule• DVB-T and DVB-T2 unquestioned• Convergence of Broadcast and cellular

31

How to prepare for the future?Some thoughts…


Today’s PMSE• Narrowband PMSE in 200 kHz does not allow for frequency diversity• Time diversity cannot be explored due to stringent latency requirements• Antenna diversity is already explored• Migration to MIMO e.g. 2x2• Massive MIMO for large events?

Economies of scale• For historic reasons PMSE uses analogue FM and Stereo as in analogue FM broadcast

with 200 kHz channel• Allows for reuse of Integrated Circuits ICs: FM Demodulator, Stereo decoder with IEM• Instead of reusing technology from Broadcast market, reuse technology from cellular

market, Basestations, handset modems (see PMSE-xG project)• Reusing cellular technology would allow for: MIMO, Wideband air interface, interference

diversity, QoS management, Macrodiversity by SFN, eMBMS, new services…

PMSE equipment vendors• Have partially distinguished themselves based on radio performance• Reusing cellular technology would leave their USP in the area of acoustic and sound

properties – acceptable?

Media convergence Cellular, Broadcast, PMSE?

6. Conclusions

32

33

Conclusions


Digital transmission• PMSE has unique demands (Latency, quality, availability, drop outs, clicks, compression,

battery operation• Robustness comes through diversity

Coexistance with other systems• With interferers having fluctuating power, peak power relevant, not average• Examples: Airband DME, LTE Duplex gap, shoulders of BS and MS

Diversity• The root of robustness with digital transmission• Time diversity cannot be explored in PMSE due to stringent latency requirement• Frequency and interference diversity not in place with todays 200 kHz narrowband PMSE • Wideband transmission desirable• Antenna diversity already in use today, however room for improvement, MIMO, massive

MIMO

Media Convergence• An option on the horizon thanks to 5G• Cellular and Broadcast already in talks on convergence through eMBMS, derivation ofa

catalogue of requirements• Cellular and PMSE already in research activities regarding necessary adaptations on 5G

standard

34

Conclusions


And the PMSE vendors?• May have to adjust their business• Radio no longer a distinguishing feature• Concentrate on sound and acoustic properties• Capsule, transducer competence is key• Uncompressed and studio quality will get a realistic option, compression technology no

longer a USP?

And the PMSE users?• Get familiar with cellular technology• Operational service, local MNO?• More diversity, more robustness• New services to be administered

Opportunities by 5G tremendous!Can it deliver what PMSE requires?

Documents

Analysis of Changes in PMSE Spectrum and Transmission ... fileAnalysis of Changes in PMSE Spectrum and Transmission Technology -Our Prediction for PMSE, Operated Under Changed Conditions