Clustering Based Multiple Hypothesis Multi-Target Localization

Cooperative detection system, which combines self organizing network and single radar system, improves the ability of cooperative localization by distributed multi-node using multi-direction scattering power of target. Multiple hypothesis (MH)-Based Algorithm for Target Localization finds all possible targets using multi-path echo information with unknown number of targets, and realizes multi-target localization by generating global hypothesis. However, a large number of targets will cause the large computational complexity of MH-Based Algorithm for Target Localization. For this problem, this paper proposes a new method of Clustering Based Multiple Hypothesis Multi-target Localization, which decomposes a large number of hypotheses into several independent clusters, and hypothesis updating, hypothesis pruning and global hypothesis generating will be implemented independently in each cluster. Compared with MH-Based Algorithm for Target Localization, the proposed method increases the computation efficiency.