Modelling LTE 800 Interference on DTT. The Approach to Mitigating Interference in the UK Marcus Buchan

Transcription

1 Modelling LTE 800 Interference on DTT The Approach to Mitigating Interference in the UK Marcus Buchan LS Summit July 2014

2 The DMSL Background DMSL was instigated by the UK Govt. in 2012 UK Govt. set-out MNO funding for DMSL DMSL uses the consumer brand at800, online as at800.tv DMSL operates under KPI s monitored by an Oversight Board Based in London using suppliers to provide key functions such as contact centre, aerial installation, mailing and marketing 2

3 The 800 MHz band plan Inverted Uplink/Downlink used in LTE800 Only 1 MHz Guard band between DTT and LTE800 3

4 4 The Interference Issue

5 Protection ratios A protection ratio is used to indicate how strong a signal needs to be to interfere with another signal. This is the ratio (in db) by which the received power of the wanted signal must exceed that of the unwanted signal to ensure that interference does not occur. For a mobile signal on an adjacent DTT channel, the protection ratio is often quoted as -25dB although can be much less if an amplifier is used -25dB 0dB Wanted DTT Signal 4G signal on adjacent channel Power Frequency 5

6 Mechanisms of Interference DTT Receiver Overload DTT Receiver Selectivity to Adjacent Channel Interference (ACLR) DTT Remote amplifier Blocking both Mast Head Amplifiers and Distribution Amplifiers 6

7 Most demanding cases A combination of some or all of the following factors may result in particularly demanding interference cases: The LTE base station(s) and the TV aerial are in close proximity There are multiple LTE base stations nearby The DTT multiplexes are on channels close to the LTE base station frequencies, particularly channel 60 The household is in a poor DTT coverage area An amplifier, or amplifiers, are used in the TV installation The householder has an indoor aerial - these are more prone to interference. The equipment quality aerial, cabling, connectors in the household is poor 7

8 Measuring a 4G Signal Display shows DTT multiplexes on Channels 51 and 52 and LTE signal at 795MHz. No interference is present on the DTT signals 8

9 Measuring a 4G Signal with LTE800 interference Display shows DTT multiplexes on Channels 51 and 52 and the LTE signal at 796 MHz after passing through an amplifier. The LTE signal level has overloaded the amplifier and caused it to operate in a non-linear fashion 9

10 Filters The primary means of mitigating interference is the use of a filter Filters fix the vast majority of interference issues when used in the right place. at800 has procured a selection of filters for different installation types and these will be available to installers 10

11 Interference Modelling Based On Ofcom Model and Protection Ratios Uses the UK Planning Model (Pre-calculated DTT coverage data) for determining the Location Probability calculated using the median and standard deviation of the received wanted and unwanted signal power* Early Software Punch supplied by Arqiva Requirement for Software that provided a GUI and more in-depth and flexible approach LS CHIRplus chosen from a range of vendors decision based on vendors experience in broadcast TV and interference analysis *Ofcom:Technical analysis of interference from mobile network base stations in the 800 MHz band to digital terrestrial television 10 th June

12 Challenges to DTT Interference Modelling Assumptions are made about the DTT receive system with regards to: Gains and Losses Angular discrimination of Receive antenna, ITU-R Rec Location of the DTT transmitter Assumptions are made regarding the Protection Ratios used: An average DTT receiver or External Amplifier is assumed enodeb spectral Mask and ACLR Assumptions are made within the UKPM Fixed Log Normal Standard deviation of 5.5dB Percentage Interference of 1% used 12

13 What does the model do? Uses MNO LTE800 network data to predict LTE800 coverage per pixel per LTE800 cell Adds the DTT Receive Antenna discrimination (polar and angular) and Protection Ratio to each LTE interferer per pixel Uses the Schwartz & Yeh Log Normal Power summation method to sum interferers per pixel Takes existing datasets of UK DTT coverage based on a 100m pixel resolution from the UK Planning model for DTT Derives degradation to Location Probability using the standard deviations of the received wanted and unwanted signal powers 13

14 Speed of Process DOES NOT REFLECT ACTUAL UK MAST DEPLOYMENTS 14

15 Analysis of Results from CHIRplus Three UKPM layers are processed by the model to determine most heavily impacted pixels using the DIA protection ratio sets Pixels are processed based on severity of degradation Households within an affected pixel are mailed (leaflet) if criteria are met Data on Households previously mailed is used to reduce over-mailing 15

16 Model Tuning Ongoing efforts to improve model accuracy include: Looking at propagation model used, currently SEAMCAT HATA Impact of clutter Impact of Terrain Accuracy of Protection Ratios Power characteristics of MNO LTE800 enodeb equipment and ACLR Looking to improve current use of UKPM data Improvement in post processing methodology 16

17 Current Status 10 months of rollout; ~ 2000 confirmed interference cases to date Mostly Block B&C active in UK so far; little adjacent channel data All confirmed cases of LTE800 interference have occurred within 900m of an LTE mast Call Centre constantly updated regarding areas of deployment and activation Call Centre Triage system used to asses if a filter should be posted/team deployed 17

18 Way Forward Further model tuning required this will help refine knowledge on the mechanisms of interference from mobile networks on DTT Likely future release of DTT 700 MHz spectrum (694 MHz to 790 MHz) larger Guard Band of 9MHz, Uplink adjacent to DTT spectrum to comply with existing Handsets 18

19 Alternatives to Modelling Reactive scenario using Call Centre to triage complaints only and issuing advice and post filters as necessary Spatial determination of proximity to mast to determine likelihood of interference Mix of both Reactive scenario and Spatial determination What do we learn about DTT interference using these alternatives? 19

20 Any Questions? Contact Details: 20