This paper proposes a novel scheme for multi-static passive radar processing, based on soft-input soft-output processing and Bayesian sparse estimation. In this scheme, each receiver estimates the probability of target presence based on its received signal and the prior information received from a central processor. The resulting posterior target probabilities are transmitted to the central processor, where they are combined, to be sent back to the receiver nodes or used for decision making. The performance of this iterative Bayesian algorithm comes close to the optimal Multi-Input Multi-Output (MIMO) radar joint processing, although its complexity and throughput are much less than MIMO radar. Also, this architecture provides a tradeoff between bandwidth and performance of the system. The Bayesian target detection algorithm utilized in the receivers is an iterative sparse estimation algorithm named Approximate Message Passing (AMP), adapted to SISO processing for passive radar. This algorithm is similar to the state of the art greedy sparse estimation algorithms, but its performance is asymptotically equivalent to the more complex l1-optimization. AMP is rewritten in this paper in a new form, which could be used with MMSE initial filtering with reduced computational complexity. Simulations show that if the proposed architecture and algorithm are used in conjunction with LMMSE initial estimation, results comparable to jointly processed basis pursuit denoising are achieved. Moreover, unlike CoSaMP, this algorithm does not rely on an initial estimate of the number of targets.

Keywords