Robust Target Localization for Distributed MIMO Radar Under Space-Time Imperfections via Approximate Belief Propagation

This letter investigates the problem of robust target localization in distributed multiple-input multiple-output (MIMO) radar systems under space-time imperfections, i.e., antenna position errors and clock biases at both transmitters and receivers, using bistatic-range (BR) measurements. The space-time errors and BR measurement noise are characterized by hierarchical Gaussian distributions with Gamma hyper-priors imposed on their precision parameters. We develop a computationally efficient approximate belief propagation algorithm to resolve the intractable loopy Bayesian inference problem. In particular, the BR-related and precision-related messages are handled through local linearization and Gaussian moment-matching projection. This enables fully Bayesian adaptive inference without exact prior variances of the space-time errors and noise. Simulation results under various conditions demonstrate the effectiveness of the proposed algorithm.