Cramer-Rao type bounds for sparsity-aware multi-target tracking in multi-static passive radar

Abstract

Sparsity-aware multi-sensor multi-target tracking (MTT) algorithms comprise a two-step sequential architecture that cascades a group sparse reconstruction scheme and a multi-target tracker. The former exploits the a priori knowledge that the measurements across multiple sensors share a common sparse support in a discretized target state space and provides a computationally efficient approach for centralized fusion of the multi-sensor information. In the succeeding step, the multi-target tracker performs data association, compensates for the missed detections, and removes the clutter components, so as to improve the accuracy of multi-target state estimates. In many practical applications, the observation suffers from a high proportion of missing samples, rendering it difficult to accurately estimate the multi-target states using the group sparse reconstruction methods. Therefore, it is of significant interest to analyze the performance loss due to missing samples. In this paper, we analytically evaluate the Cramer-Rao type performance bounds for the sparsity-aware multi-sensor MTT algorithms in a multi-static passive radar system and evaluate the performance loss due to missing samples in the measurement vectors.