Automotive Radar Target Detection in Widely Separated and Distributed Aperture Radar Systems
Moein Ahmadi, Björn E. Ottersten, Bhavani Shankar M. R., Thomas Stifter
Abstract
This paper presents an approach to target detection in automotive radar systems, where the highly dynamic nature of the sensor platform and environment, along with challenges such as hardware cost and installation constraints, necessitates a general sensor configuration that integrates widely separated, colocated MIMO, and distributed aperture radar (DAR). A joint Doppler processing and MIMO transmit demodulation technique is proposed, utilizing arbitrary DDM, TDM, or BPM precoding matrices with a lower-dimensional antenna steering matrix as the detection input. The signal model incorporates environmental information, such as the cell under test (CUT) range and line-of-sight regions, to enhance detection performance. A distributed Generalized Likelihood Ratio Test (GLRT) detector is derived using secondary data with CFAR property, and the robustness of the proposed detectors is evaluated under mismatch conditions using mesa plots. Simulation results demonstrate the effectiveness of the proposed approach, evaluated using key performance metrics such as probability of detection, probability of false alarm.
BibTeX
@inproceedings{icassp2025_automotiveradart,
title = {Automotive Radar Target Detection in Widely Separated and Distributed Aperture Radar Systems},
author = {Moein Ahmadi and Björn E. Ottersten and Bhavani Shankar M. R. and Thomas Stifter},
booktitle = {ICASSP 2025},
year = {2025}
}