Data Association and Multi-Target Localization Using Particle Swarm Optimization

In this paper, we propose a novel method for data association and multi-target localization. A 3-stage approach, Optimization-Association-Localization, is proposed which can handle false alarms and miss detections in measurements. In this regard, as the first stage, we use arithmetic-geometric mean inequality to convert ML solution to a novel optimization problem, the solution of which gives a rough estimate of targets' locations. Then, using the nearest line method, we associate data and utilize the single-target localization algorithm to obtain final locations for targets. AOA-based localization which is one of the most straightforward passive localization methods to implement, is considered in this paper for evaluation of the proposed method. Simulation results show that the proposed approach can solve the multi-target localization problem accurately and CRLB can be attained.