Efficient Multi-Target Localization Through Geometric Data Filtering in Passive Multistatic Radar Systems
David Melon Fuksman, Nahuel Almeira, Octavio Cabrera Morrone · 2024
We introduce a novel technique for increasing the efficiency of localization algorithms in multistatic passive radar systems. Our method consists of filtering out data that fall outside the space of feasible bistatic range measurements using easy-to-implement geometric constraints. This strategy successfully eliminates faulty measurements in single-target scenarios without compromising the localization accuracy. In multi-target scenarios, it provides a substantial speedup, increasing with the number of targets, still without sacrificing localization accuracy. This makes the filtering method particularly advantageous for real-time applications with many simultaneously observed targets. We support our results on extensive simulations, including different multistatic configurations, and discuss the filter's performance in a DVB-T-based passive radar.