MmWave for Vehicular Communications
The use of very large antenna arrays composed of tens to hundreds of elements at mmWave frequencies is necessary to maintain sufficient link quality. Such antenna arrays are feasible due to the small antennas at these high frequencies. However, beam alignment of arrays at this scale is challenging due to the overhead in determining the optimal transmit and receive beams. Existing beam alignment schemes proposed for mmWave applications are largely based on beam sweeping and make use of hierarchical beam codebook for efficient algorithm design, which requires the use of wide beams at the start. Wide beams are more susceptible to Doppler spread, which is another challenge for beam sweeping based method. It should be noted that these solutions do not make use of any side information. Devices with positioning capability are becoming more and more ubiquitous, e.g. in the form of portable navigation devices and smart phones. Motivated by this fact, we will develop a low overhead beam alignment algorithm leveraging vehicle’s position as side information using a learning based approach. We will first focus on vehicular to infrastructure (V2I) communications.
We recently have studied the potential of leveraging position information to switch the beams and eliminate the need for beam sweeping. Our result shows that mmWave with beam switching has the potential of providing multi-Gbps communications in vehicular environments. Position prediction accuracy is crucial for the success of the beam switching. Our analysis shows that there exists a fundamental tradeoff in the design of beamwidth: wider beams suffer from insufficient power but narrower beams are more sensitive to position estimation error. We also study the fundamental limit of mmWave systems in vehicular environments. In particular we study the connection between the channel coherence time and the beamwidth. In our study, we have proposed a new channel model based on the one-ring model that can capture the pointing error due to the receiver’s motion. With this proposed model we showed that there is an optimum operating beamwidth that is non-zero and maximizes the coherence time in contrast to existing models that predict infinite coherence time as the beamwidth approaches zero.
This research is partially supported by the U.S. Department of Transportation through the Data-Supported Transportation Operations and Planning (D-STOP) Tier 1 University Transportation Center and by the National Science Foundation under Grant No. NSF-CCF-1319556.
V. Va and R. W. Heath, Jr., “Basic Relationship between Channel Coherence Time and Beamwidth in Vehicular Channels”, to appear in Proc. of the IEEE 82nd Vehicular Technology Conference (VTC2015-Fall), Boston, USA, September 6-9, 2015.
V. Va, X. Zhang, and R. W. Heath, Jr., “Beam Switching for Millimeter Wave Communication to Support High Speed Trains”, to appear in Proc. of the IEEE 82nd Vehicular Technology Conference (VTC2015-Fall), Boston, USA, September 6-9, 2015.
Data-Supported Transportation Operations & Planning Center A Tier 1 USDOT University Transportation Center at The University of Texas at Austin