Applying neural network algorithm to data association technique

Data association plays an important role in radar tracking algorithm. The problem of tracking multiple targets is studied in this paper. In order to solve the complicated situation and reduce computation burden because of the multiple tracking environment, an approach has been developed in this paper. This algorithm is implemented with an adaptive filter which consists of a data association technique denoted competitive Hopfield neural network and Kalman filtering to solve both data association and target tracking problems simultaneously. In order to prove the tracking performance, a computer simulation algorithm is proposed in this paper. Because of its computation capability of this algorithm, the radar measurement related to existed target tracks can be chosen optimally. Computer simulation results indicate that this approach successfully and optimally solves the data association problems.