Knowledge-Aided Bayesian Detection of Distributed Target for FDA-MIMO Radar in Gaussian Clutter

For Frequency diverse array multiple-input multiple-out (FDA-MIMO) radar, this paper studies the knowledge-aided Bayesian detection for a one-range-bin distributed target with multiple scatters operating in Gaussian clutter environment with unknown and stochastic clutter covariance matrix. Specifically, we build the FDA-MIMO receive signal model by capitalizing on orthogonality in the frequency domain. Subsequently, an inverse complex Wishart distribution is assigned to the clutter covariance matrix for mathematical tractability, serving as knowledge-aided information. With free training data, two adaptive detectors are introduced by leveraging the Bayesian framework, based on Rao and Wald criteria, namely, Bayesian Rao (BRao) and Bayesian Wald (BWald), respectively. Notably, it is essential to highlight that the received FDA-MIMO signals can be directly applied to adaptive detectors without needing matched filtering. The simulation results confirm that, in the case of signal matching, the BWald can provide detection performance comparable to that of the existing BGLRT. Additionally, when facing mismatched signals, the proposed BWald and BRao detectors demonstrate stronger robustness and selectivity capabilities.