Evaluation of Twin Passive Linear Array Sonar TDOA Estimation for Ghost Track Elimination

Tracking multiple targets using multiple sensors often poses a challenging measurement association problem. Many methods have been proposed for multi-target tracking using several fixed passive hydrophone arrays. A key element of all these methods is minimizing the impact of ghost targets, which are the result of incorrect measurement assignments. Leveraging a sea trial campaign involving three boats with known positions and two linear arrays, we explore the potential of using signal coherence between the two arrays to tackle this problem. Our results are promising, demonstrating that the time difference of arrival between the signals measured by the two arrays can be successfully estimated in most instances if the relative Doppler shift between the signals is considered in the estimation process. Importantly, we show that ghost targets are rarely misclassified as real targets, 2% false positives, thereby validating the efficacy of our proposed method.