All neighbor fuzzy relational data association for multitarget tracking in the presence of ECM

Abstract

This paper proposes a novel data association approach based on fuzzy relational clustering for multi target tracking in the presence of electronic counter measures (ECM). Likelihood values and similarity index are calculated for each observation obtained from radar. Expectation maximization technique is applied to obtain possibility association matrix. Simulation results demonstrate that, proposed method performs better, when compared to conventional joint probability association (JPDA) and fuzzy clustering (FCM) approaches in terms of position and velocity root mean square error (RMSE). Further, current approach yielded average reduction of 50.5% and 35.5% for position and velocity RMSE values respectively in case of linear crossing targets.