Transmitter Subset Selection in FM-Based Passive Radar Networks for Joint Target Parameter Estimation

Passive radar network systems utilize multipletransmitters of opportunity and multichannel receivers to offerremarkable performance improvement due to the advantage ofsignal and spatial diversities. The frequency modulation (FM)commercial radio signals have become attractive for passiveradar applications owing to their wide-spread availability andthe favourable Doppler resolution. In this paper, two transmittersubset selection schemes, balancing the trade-off between targetparameter estimation accuracy and infrastructure utilization,are proposed for FM-based passive radar networks. In thefirst, the subset size of selected transmitters employed in theestimation process is minimized by effectively selecting a subset oftransmitters such that the required target parameter estimationmean-square error (MSE) threshold is attained. In the second,an optimal subset of transmitters of a predetermined size is selected such that the estimation MSE is minimized. Theseproblems are formulated as a knapsack problem (KP), wherethe coherent Cram´er-Rao lower bound (CRLB) is used as a performancemetric. Both transmitter subset selection schemes aretackled with greedy selection algorithms by successively selectingtransmitters so as to minimize the performance gap between theCRLB and a predetermined MSE threshold or a predeterminedsubset size. Numerical simulations demonstrate that the problemof transmitter subset selection is not only a function of thetransmitted waveforms but also of the relative geometry betweenthe target and the passive radar network systems, which leadsto possible reductions in both communication links and signalprocessing costs.