Robust Target Localization for Multistatic Passive Radar Networks

Passive multistatic radar networks localize a noncooperative target exploiting the bistatic measurements associated with signals emitted by several transmitters of opportunity and acquired by multiple receivers. However, in realistic scenarios, it might happen that one or more receivers are damaged and/or under malicious attacks with a consequent degradation of the system performance. This letter proposes a procedure to reveal sensors that return anomalous measurements due to an attack or a failure. Specifically, we first detect such measurements by solving a binary hypothesis test; then, we cancel them and estimate the final delays as well as the target position. The performance of the overall architecture is assessed using both synthetic data and real-recorded data.