Adaptive and Fast Combined Waveform-Beamforming Design for MmWave Automotive Joint Communication-Radar

High data rate transmission and high-resolution radar detection for applications like autonomous driving will be made possible by millimeter-wave (mmWave) joint communication radars (JCR). However, because of the directional communication beam used, previous JCR systems based on mmWave communications hardware have a small angular field-of-view and poor radar estimating accuracy. In this study, we propose a phased-array architecture-based adaptive and fast combined waveform-beamforming design for the mmWave automobile JCR that allows for a trade-off between radar performance and communication. Our JCR design uses two-dimensional compressed sensing (CS) in the space-time dimension and circulant shifts of the transmit beamformer to obtain radar channel measurements in order to quickly estimate the mmWave automobile radar channel in the Doppler-angle domain with a wide field-of-view. While keeping in mind the space-time sampling limitations of our situation, we optimize these circulant shifts to reduce the coherence of the CS matrix. A normalized mean square error (MSE) metric for radar estimating and a distortion MSE metric for data communication—which is comparable to the distortion metric in the rate-distortion theory—are used to assess the JCR performance trade-offs. Furthermore, for the adaptive JCR coupled waveform-beamforming design, we formulate a weighted average optimization problem based on mean square error. Numerical results show that, at the cost of a slight decrease in the communication distortion MSE, our suggested JCR design allows for the estimate of short- and medium-range radar channels in the Doppler-angle domain with a low normalized MSE.

Keywords: Automotive radar, millimeter-wave vehicular communication, joint communication-radar, partial Fourier compressed sensing, adaptive waveform and beamforming design.