A Riemannian-Based Joint Design Framework of Mimo Radar Transmit Waveform And Receive Filter Via Information Theory.

Jie Li,Yan Huang,Qihui Wu,Arye Nehorai
DOI: https://doi.org/10.1109/ICASSP48485.2024.10446363
2024-01-01
Abstract:In this paper, we explore the joint design of a transmit waveform and receive filter to enhance the detection performance of multiple-input multiple-output (MIMO) radar. Target echoes are assumed to be embedded in signal-dependent interference and colored Gaussian noise. As design metrics, we exploit two information-theoretic criteria, including mutual information (MI) and relative entropy. The joint design problems of MIMO radar associated with different information- theoretic criteria are established as a unified optimization framework within a constant-envelope (CE) constraint. We propose an efficient method based on the Riemannian optimization framework, which transforms the constraint optimization problems into unconstrained problems by leveraging the geometry of the feasible region. Several numerical examples are included to demonstrate the effectiveness of the proposed method.
What problem does this paper attempt to address?