Vehicular mmwave Communication: Opportunities and Challenges Professor Robert W. Heath Jr., PhD, PE Wireless Networking and Communications Group Department of Electrical and Computer Engineering The University of Texas at Austin Thanks to sponsors including the US Department of Transportation, the Texas Department of Transportation, Toyota IDC, and National Instruments www.profheath.org
UT is a collaboration hub for wireless + transportation 2
Transportation and wireless @ UT Austin Robert W. Heath Jr. (2015) SMART TRANSPORTATION WIRELESS COMMUNICATION TRAFFIC MODELING SIGNAL PROCESSING POLICY AND PLANNING MACHINE LEARNING & BIGDATA u WNCG brings renowned wireless expertise & deep industry connections u CTR has more than 50 years of transportation research experience UT is well positioned to develop wireless networks for transportation systems
A new transportation initiative at UT Austin Robert W. Heath Jr. (2015) u SAVES: Center for Situation-Aware Vehicular Engineering Systems u Focuses on wireless, networking, and sensing challenges in vehicular systems u Brings together stakeholders in different aspects of the automotive market SAVES is built around TOYOTA as a founding member, lookging for partners 4
Why vehicular communications at mmwave? 5
5G will enable key changes in transportation AUTOMATED CAR Robert W. Heath Jr. (2015) AUTONOMOUS DRIVING ROAD SAFETY TRAFFIC EFFICIENCY u Automated cars needs sensing capabilities ª Increased number of sensors in a vehicle ª Vehicular communications to share sensing data and improve sensing capability New challenges for the underlying communication system *5G-PPP White Paper on Automotive Vertical Sector, October 2015, https://5g-ppp.eu/white-papers/ 6
DSRC: current technology for vehicular communications u u u u Forward collision warning, do not pass warning, blind intersection warning, etc. Non-safety apps also possible improve congestion, weather, toll collection Based on IEEE 802.11p, IEEE 1609.x, SAE standards Supports very low data rates (27 Mbps max, much lower in practice) DSRC is expected to be mandated for all vehicles in US by NHTSA by 2017 *NHTSA, Vehicle-to-Vehicle Communications: Readiness of V2V Technology for Application, Aug. 2014 **John B. Kenney, DSRC: Deployment and Beyond, WINLAB presentation, May 2015. 7
Massive data rates from sensors GPS Lidar Videocameras Each sensor generates data Infrared cameras Ultrasonic sensor Inertial motion sensor Automotive radar u Connected vehicle is expected to drive 1.5GB monthly mobile data in 2017 ª May be handled with a combination of conventional cellular and DSRC u Autonomous vehicle can generate up to 1 TB of data in a single trip ª 4G and DSRC can not support these data rates *http://low-powerdesign.com/sleibson/2011/05/01/future-cars-the-word-from-gm-at-idc s-smart-technology-world-conference/ **Cisco, The Internet of Cars: A Catalyst to Unlock Societal Benefits of Transportation, Mar. 2013 ***http://www.sas.com/en_us/insights/articles/big-data/the-internet-of-things-and-connected-cars.html Lots of sensors in a vehicle Massive amount of data per vehicle MmWave is the only viable approach for high bandwidth connected vehicles 8
Millimeter wave spectrum for 5G <1 GHz 5.2 GHz 555 MHz 2.4 GHz 100 MHz 900 MHz 2.6 MHz CmWave 28 GHz 1.3 GHz MmWave 39 GHz 37/42 GHz 1.4 GHz 2.1 GHz 20 GHz x100 GHz 60 GHz 7 GHz E-band 10 GHz total 6 GHz Unlicensed 100 GHz Even more spectrum More spectrum, in bands not previously used for cellular * T. Rappaport et al., Millimeter wave mobile communications for 5G cellular: It will work! IEEE Access, 2013. ** W. Roh et al., "Millimeter-wave beamforming as an enabling technology for 5G cellular communications: theoretical feasibility and prototype results,, IEEE Commun. Mag.,, 2014 *** A. Osseiran et al.,"scenarios for 5G mobile and wireless communications: the vision of the METIS project," iieee Commun. Mag., May 2014 9
Implications of using mmwave in automotive Robert W. Heath Jr. (2015) Increased sensing capability in the car Joint automotive radar and communication is possible Cloud driven autonomous driving Sensing technologies can be used to help establishing communications 10
Key features for mmwave vehicular communications 11
MmWave vehicular communications Vehicle driving cloud blockage mmwave base station V2I communication beam directional beamforming V2V communication beams MmWave vehicular communications introduce new challenges 12
Implications of directional beamforming in V2X Beam training overhead is critical due to high mobility Optimum beamwidth has to be selected Multiuser operation strategies New channel models for V2V and V2I are needed 13
Selecting the optimum beamwidth u Mathematical expression relating coherence time and beamwidth ª Accounts for beam pointing angular difference as oppose to classical models u Optimum beamwidth is a tradeoff between pointing error and Doppler Beams should be narrow but not too pointy *Vutha Va, and Robert W. Heath, Jr, "Basic Relationship between Channel Coherence Time and Beamwidth in Vehicular Channels,'' IEEE Vehicular Technology Conference (VTC 2015-Fall), 2015. 14
Defining the coherence time for beam alignment Robert W. Heath Jr. (2015) Four channel directions u u u Received power The preferred channel direction changes much more slowly than the instantaneous fading envelope Beam coherence time relates to changes in the mean direction Long term beamforming can be used Overheads of beam training are much less significant than expected *V. Va, J. Choi, and R. W. Heath Jr. The impact of beamwidth on temporal channel variation in vehicular channels and its implications. arxiv preprint. 15
Using position information to reduce beam alignment overhead in mmwave V2X u Each vehicle decides candidate beams from other vehicles position and size info DSRC modules or automotive sensors can be used to reduce overhead Junil Choi, Nuria Gonzáalez-Prelcic, Robert Daniels, Chandra R. Bhat, and Robert W. Heath Jr, Millimeter Wave Vehicular Communication to Support Massive Sensing, to be submitted December 2015 16
Adding radar to the infrastructure for multiuser operation radar mmwave BS popsci.com u A BS with a radar can capture information of the scattering environment u Used to design multiuser beamforming, support remote car traffic control Sensing at the infrastructure can help in establishing the communication links 17
Channel models for mmwave vehicular Typical antenna height: 1.5 m u Channel classifications considered at 5.9GHz is unlikely to scale ª More sensitive to antenna orientation ª More sensitive to traffic density (higher blockage probability) ª Effect of directive transmission is unknown u Few measurements available * T. S. Rappapport, R. W. Heath Jr., R. C. Daniels, and J. N. Murdock, Millimeter Wave Wireless Communications, Pearson Prentice-Hall, 2014 ** S. Takahashi, et al., Distance dependence of path loss for millimeter wave inter-vehicle communications, in VTC 2003-Fall, Oct. 2003, ** W. Schafer, Channel modelling of short-range radio links at 60 GHz for mobile intervehicle communication, in IEEE 41st VTC, May 1991. 18
Defining new channel models Persistence process for each cluster modellating blockage Doppler shift for each multipath component Delay of cluster c and ray p θ AoA angle spread Steering vectors Modified Sen-Matolak model* Clustered in space: intra-cluster structure u New models must account for mmwave propagation characteristics u Channels are sparse in the angular domain, a few paths exist u Blockage has to be introduced in the channel model * V. Va, N. Gonzalez-Prelcic and R. W. Heath Jr., A cluster-based approach for vehicular channel models at millimeter wave, in preparation. 19
Implications of blockage in V2X Automotive sensors can be used to predict blockage Antenna diversity to overcome blockage Multibeam transmission to overcome blockage Blockage can help to mitigate interference 20
Antenna placement and blockage u A classic problem even at low frequencies* ª Shadowing becomes blockage for mmwave ª Directional transmission adds another challenge u V2V require 360 degree coverage but antennas can not penetrate car ª Front bumper location causes blockage at the back side ª Rooftop location causes blockage at the front side due to roof curvature ª Sensitive to antenna orientation MmWave propagation add new challenges to antenna placement * C. Mecklenbrauker, et al., Vehicular channel characterization and its implications for wireless system design and performance, Proc. of the IEEE, July 2011. 21
Antenna diversity to overcome blockage in V2I u u A BS with a radar is assumed at the infrastructure side ª Antennas are assumed to be placed at the virtual scattering points in the car ª Radar info is used to design a multi-beam pattern to track several antennas High mobility is considered and the positions of the antennas are predicted Sensing at the infrastructure can help to manage blockage * N. González-Prelcic, R. Méndez-Rial and R. W. Heath Jr., Radar-aided multibeam directional beamforming for mmwave vehicle to infrastructure communications, in preparation. 22
Predicting blockage from out-of band sensing Robert W. Heath Jr. (2015) u A BS equipped with a radar is assumed u Radar and communication operate at different mmwave bands u Radar can detect potential obstacles and their associated mobility u Machine learning can classify particular radar responses as blockages Sensing & learning are symbiotic technologies Junil Choi, Nuria Gonzáalez-Prelcic, Robert Daniels, Chandra R. Bhat, and Robert W. Heath Jr, Millimeter Wave Vehicular Communication to Support Massive Sensing, to be submitted December 2015 23
Joint vehicular communication and radar 24
Combining communication and radar at mmwave Emergency Van Robert W. Heath Jr. (2015) Communication Signal Direction of Cruise Radar Multi-beam Emergency Event u MmWave radars widely used in automotive safety applications ª Expensive: automotive radar modules cost around $1,500 ª Easy to get spoofed compared to communication u Why not share common equipment with communication? ª Combines the objectives of radar and communication Shared hardware reduces cost/size and makes efficient spectrum usage 25
MmWave communication-radar challenges Robert W. Heath Jr. (2015) u Optimization of sensing and data communication ª LFM # waveform provides low data rate ª DSSS # exhibits poor radar performance ª No single waveform yet available ª Interference issue u Assumption of full-duplex ª Separate transmit and receive antenna ª Use directional antennas Communication-radar (RadCom) Application Scenario Many open problems to design joint communication and radar systems LFM # : Linear frequency modulated waveform, which is a radar waveform DSSS # : Direct spread spectrum, which is a communication waveform *L. Han and K.Wu,``Joint wireless communication and radar sensing systems-state of the art and future aspects,'' IET Microwaves, Antennas & Propagation, vol. 7, no. 11, pp. 876-885, 2013. 26
Joint automotive radar and communications based on 802.11ad u u IEEE 802.11ad mmwave waveform works well for radar ª Special structure of preamble enables good ranging performance ª Leverages existing WLAN receiver algorithms for radar parameter estimation Target vehicle information from 11ad radar can be directly used for communication Joint system provides safety capabilities at lower cost *P. Kumari, N. González Prelcic and R. W. Heath, Jr, ``Investigating the IEEE 802.11ad Standard for Millimeter Wave Automotive Radar,'' IEEE VTC-Fall, 2015. 27
Prototyping and measurements 28
UHF radar and mmwave communication Robert W. Heath Jr. (2015) Transmit 802.11 OFDM waveform (DSRC) through single forwarddirectional antenna at 5.9 GHz 802.11 RF/Analog UT custom DSP 802.11 Baseband UT custom DSP u UT custom DSP separates simultaneous TX/RX operations u It provides meter-level range of largest target Platform can be used to obtain out-of-band information for communication at mmwave 29
UT mmwave communication and radar prototype National Instruments PXI Chassis Interfaces with custom RF Communication transmitter and radar receiver Steerable, Multi-level Scanning Communication receiver Processing in NI FPGA module NI Signal Generator (TX) Custom RF TX Antenna RX Antenna Custom RF LRR SRR MRR NI Digitizer (RX) Processing in NI FPGA module Processing in NI FPGA module NI Digitizer (RX) Custom RF FPGA for real-time processing RX Antenna Platform allows V2V communication link prototyping, new radar waveform development, and joint communication and radar waveform testing 30
Multi-sensor measurement campaign National Instruments PXI Chassis Interfaces with USRP-RIO & ESR Cohda Wireless DSRC unit National Instruments USRP-RIO w/ 5.8 GHz RF, 802.11p PHY+MAC RX TX RX Delphi ESR 2.5 77 GHz RADAR L-COM 23-dBi HG4958-23P Patch - 2 receive (RX) - 1 transmit (TX) u Joint measurements using the WiFi-based radar, off-the-shelf mmwave radar, GPS, and DSRC Collect sensing data in different bands and asses limitations of different technologies 31
Conclusions 32
Conclusions Opportunities Millimeter wave is ripe for V2X applications Natural synergies with automotive radar, circuit design, etc. Only feasible way to enable connected + autonomous Challenges Unlicensed band, co-exist with radar, new spectrum? Complete redesign of physical layer required, mobility Support of full duplex for more complicated waveforms Antenna mounting, packaging, cabling 33