TV White Spaces characterization of DVB T systems Rogério Dionísio Instituto de Telecomunicações (Portugal) STG Workshop, October 5, 2010 Page 1
Contents COGEU Project overview Simulations Case study 1: Interference from LTE BS / UE to DTT Case study 2: Interference from Cognitive LTE BS / UE to DTT Case study 3: Interference from LTE BS / UE to PMSE Final comments Conclusions Future work STG Workshop, October 5, 2010 Page 2
At a glance: COGEU STREP Project (www.ict-cogeu.eu) Duration: Jan 2010 Dec 2012 9 partners STG Workshop, October 5, 2010 Page 3
COGEU: System Requirements Protection of DVB-T: Geolocation Database The system should be able to access geo-location database to retrieve spectrum availability information (TVWS pool). The most important is that the database stores the required information to compute the TVWS spectrum pool available in a specific location. Information such as DVB protected areas; specifications of DVB transmitters, advanced propagation models, protection rules, can be used to compute the maximum transmit power. Protection of PMSE: Spectrum Sensing Autonomous sensing is mandatory for PMSE and optional for DVB-T signals (mainly protected by the geolocation database). Detection thresholds are adopted from current regulatory framework. The system should provide a signaling channel for reporting of local sensing data and supports centralized cooperative sensing. STG Workshop, October 5, 2010 Page 4
How SEAMCAT could be used? Parameters from broadcasters and regulation TVWS maps computation based on SEAMCAT Channel X DVB-T Tx location Antenna Height Operating channel Transmitted power (EIRP) Emission masks Radiation pattern Protection requirements Minimum strength field=41 dbuv/m Co-channel protection D/U=23 db Adjacent channel protection D/U=-28 db TV set mask Resolution grid 1% interference tolerance Channel Y Populating the database Geolocation Database Max Power [dbm] Location TVWS spectrum pool Database access COGEU BROKER STG Workshop, October 5, 2010 Page 5
Contents COGEU Project overview Simulations Case study 1: Interference from LTE BS / UE to DTT Case study 2: Interference from Cognitive LTE BS / UE to DTT Case study 3: Interference from LTE BS / UE to PMSE Final comments Conclusions Future work STG Workshop, October 5, 2010 Page 6
Victim link DVB-T transmitter: Frequency: 658 MHz Power: 79.15 dbm Antenna height: 200 m antenna type: 0 dbi omnidirectional Antenna azimuth and elevation aligned with DVB-T receiver antenna. drss STG Workshop, October 5, 2010 Page 7
Victim link DVB-T receiver: Antenna height: 10 m Antenna peak gain: 9.15 dbi Antenna horizontal pattern Noise floor: -98.17 dbm Sensitivity: -77.17 dbm Reception bandwidth: 7,610 MHz Blocking response: User-defined (Attenuation) STG Workshop, October 5, 2010 Page 8
Victim link Propagation model: ITU-R P.1546 general environment: Rural system: Digital (Bw > 1 MHz) Relative location between DVB-T transmitter and receiver is 52.9 km. DVB-T receiver is located at the edge of the cell coverage, where mean received power is -68 dbm (56.21 dbμv/m @ 10m) drss 52.9 km -68 dbm 79.15 dbm STG Workshop, October 5, 2010 Page 9
Interfering link LTE BS transmitter: Frequency: 658 MHz Antenna height: 10 m Antenna: tri-sectored directional EIRPmax: 56 dbm Spectral emission mask (5 MHz BW): LTE UE transmitter: Frequency: 658 MHz Antenna height: 1.5 m Antenna: Omnidirectional EIRPmax: 23 dbm Spectral emission mask (5 MHz BW): STG Workshop, October 5, 2010 Page 10
Interference scenario Cognitive Radio (CR) device operate outside protected contour of TV station. drss -84 dbm P[dBm] CR DVB-T transmitter 200 W (UHF)? How far away does a CR device transmitting P[dBm] need to be to ensure no harmful interference? STG Workshop, October 5, 2010 Page 11
Interference scenario For each Monte-Carlo trial: Move the relative position (x,y) of the LTE BS / UE around the DVB-T receiver Propagation model LTE transmitter DVB-T receiver (irss): Extended Hata: Above roof for BS Below roof for UE LTE Transmitter Co-channel interference drss irss Variable distance y Variable distance x STG Workshop, October 5, 2010 Page 12
Interference criteria C/N = 21 db Considering desensitization: (N+I)/N db = 0.5 db (N + I)/N = I/N + 1 = 1.122, I/N db = 10log((N+I)/N 1) = 10log(0.122) = -9.136 db, C/I db = C/N db - I/N db = 21 - (-9.136) = 30.136 db C/(N + I) db = C/I db - (N + I)/I db = C/I db - (N + I)/N db + I/N db = 20.5 db STG Workshop, October 5, 2010 Page 13
Simulation results Maximum EIRP emitted by LTE BS for 1% probability of interference with DVB-T Rx dbm 16 km radius Power (dbm) DVB-T Rx cell coverage edge DVB-T Tx STG Workshop, October 5, 2010 Page 14
Simulation results Maximum EIRP emitted by LTE UE for 1% probability of interference with DVB-T Rx dbm 4.5 km radius Power (dbm) cell coverage edge DVB-T Rx DVB-T Tx STG Workshop, October 5, 2010 Page 15
Contents COGEU Project overview Simulations Case study 1: Interference from LTE BS / UE to DTT Case study 2: Interference from Cognitive LTE BS / UE to DTT Case study 3: Interference from LTE BS / UE to PMSE Final comments Conclusions Future work STG Workshop, October 5, 2010 Page 16
Interference scenario Change the relative position (x,0) of the LTE interferer (BS or UE) from the DVB-T receiver. srss drss LTE Transmitter irss Variable distance x STG Workshop, October 5, 2010 Page 17
Victim link DVB-T Transmitter: Frequency range: from 646 to 670 MHz (8 MHz steps 3 channels) Spectral emission mask: STG Workshop, October 5, 2010 Page 18
Interfering link LTE (BS or UE) Transmitter: Unwanted emission floor: -63.8 dbm Sensing parameters: Detection Threshold: -123 dbm Probability of failure: 0 % Sensing bandwidth: 8 MHz Hidden Node Margin: LTE BS: 0 dbm LTE UE: 35 dbm EIRP max in-block limit (dbm): co-channel not allowed STG Workshop, October 5, 2010 Page 19
Simulation results LTE EIRP for 1% probability of interference Average EIRP from 3 channels Autonomous sensing is more conservative for longer distances STG Workshop, October 5, 2010 Page 20
Contents COGEU Project overview Simulations Case study 1: Interference from LTE BS / UE to DTT Case study 2: Interference from Cognitive LTE BS / UE to DTT Case study 3: Interference from LTE BS / UE to PMSE Final comments Conclusions Future work STG Workshop, October 5, 2010 Page 21
Victim link Wireless Microphone: Frequency: 658 MHz Power: 10 dbm Antenna height: 1.5 m antenna type: λ/4 or λ/2 dipole 0 dbi Omni directional in azimuth and directional in elevation Antenna azimuth and elevation aligned with PMSE receiver antenna. STG Workshop, October 5, 2010 Page 22
Victim link PMSE receiver: Blocking response: Sensitivity function Noise floor: -115 dbm Sensitivity: -110 dbm Reception bandwidth: 200 khz Antenna height: 1.5 m Antenna peak gain: 2.15 dbi Antenna: Omnidirectional STG Workshop, October 5, 2010 Page 23
Victim link Propagation model: Extended Hata SRD general environment: Urban Local environment: Indoor Propagation environment: Below roof Relative location between PWMS transmitter and receiver is 100 m. PWMS receiver is located at the edge of the cell coverage, where mean received power is -95 dbm (Squelch Threshold) 10 dbm drss -95 dbm 100 m STG Workshop, October 5, 2010 Page 24
Interference criteria C/N = 21 db Sensitivity = Noise Floor + C/(N + I) db C/(N + I) db = 95 + 115 = 20 db (N + I)/N db = C/N db C/(N + I) db = 1 db irssmax limit is - 121 dbm at the PMSE receiver I/N db = 10log((N+I)/N 1) = 10log(0.26) = 5.87 db, C/I db = C/N db I/N db = 21 ( 5.87) = 26.87 db STG Workshop, October 5, 2010 Page 25
Simulation results (Batch Operation) Co-channel Interference: Victim and interfering link: 658 MHz Increase the distance between Cognitive Radio (CR) transmitter and PMSE receiver in straight line. CR characteristics based on LTE. 10 dbm drss 100 m -95 dbm irss variable distance d (100, 300 m)? dbm -80-90 d = 100 m d = 300 m 80 70 d = 100 m d = 300 m irss (dbm) -100-110 -120-130 -121 dbm Probability of interference (%) 60 50 40 30 20 19 % -140 10-150 -15-10 -5 0 5 10 15 20 Interferer EIRP (dbm) 0-15 -10-5 0 5 10 15 20 Interferer EIRP (dbm) STG Workshop, October 5, 2010 Page 26
Simulation results (Batch Operation) Adjacent-channel Interference: Victim link: 658 MHz Interfering link: 666 MHz (n + 1), 674 MHz (n + 2) and 682 MHz (n + 3) 10 dbm drss -95 dbm irss? dbm 100 m Variable distance (30-30 m) -80-85 -90 1 channel separation 2 channel separation 3 channel separation -95-100 irss (dbm) -105-110 -115-120 -121 dbm -125-130 -135 30 50 100 150 200 250 300 Separation distance (m) STG Workshop, October 5, 2010 Page 27
Contents COGEU Project overview Simulations Case study 1: Interference from LTE BS / UE to DTT Case study 2: Interference from Cognitive LTE BS / UE to DTT Case study 3: Interference from LTE BS / UE to PMSE Final comments Conclusions Future work STG Workshop, October 5, 2010 Page 28
Conclusions TVWS calculation (geolocation database approach) LTE BS interferes in DTT reception more than LTE UE. Autonomous sensing Cognitive devices with more restrictive power emissions than with the geolocation database approach. Impact of wireless microphones on TVWS maps co-channel interference impose severe limitations (LOS) adjacent channel interference is viable outside a protection distance from PMSE devices. Potential of SEAMCAT as a simulation tool Simulate several victim links at once? STG Workshop, October 5, 2010 Page 29
Future work TVWS characterization Autonomous sensing for wireless microphones indoor, outdoor, with or without LOS, etc srss drss irss SEAMCAT Post-processing plug-in to improve simulation time Avoid repetitive tasks done by the user (batch processing) and by external tools (Ex: Automator) Propagation model plug-in for wireless microphones STG Workshop, October 5, 2010 Page 30
Acknoldgement The research leading to these results has received funding from the European Community s Seventh Framework Programme [FP7/2007-2013] under grant agreement No. 248560 [COGEU]. STG Workshop, October 5, 2010 Page 31
Thank you! Questions? STG Workshop, October 5, 2010 Page 32
STG Workshop, October 5, 2010 Page 33