Waveform design for target classification in MIMO radar systems

MIMO Radar sensors are relatively immune to environmental conditions and can work in complete darkness and also bright daylight. Also, because of ultra-high range resolution (ability to discriminate closely spaced objects) and sensitivity (ability to detect weak signals), these sensors are ideal for different applications, including healthcare and security, provided that appropriate transmit and receive strategies are employed in the sensor. Because of their high frequency, with appropriate signal processing, these sensors potentially can classify and sometimes identify different objects. In this paper, we propose a waveform design framework to enhance target classification performance for the MIMO radar systems. To this end, we enhance object discrimination by maximizing the differences between received signal from every target. We show that this difference depends to the radar transmit waveform. Thus, we design the probing signal, by solving a non-convex optimization problem while considering unimodular/discrete phase, and similarity constraints in the design stage. To solve the optimization problem, we propose Coordinate Descent (CD) framework, and iteratively design the code entries. The results show that the performance of the proposed framework enhances the discrimination performance for automotive MIMO radar systems.(c) 2022 Elsevier Inc. All rights reserved.

Automated summary

In this paper, a waveform design framework is proposed to enhance target classification performance for MIMO radar systems. Object discrimination is enhanced by maximizing the differences between received signals from every target.