Application of Neural Network data associating method in the Radar Network system

For data association problem during the multiple targets tracking in the Radar Network system, we proposed a kind of improved Discrete Hopfield Neural Network (DHNN) data associating method. We used the parallel computation and optimization ability of DHNN, and improved the global optimization ability of DHNN by modifying the energy function of DHNN and applying the simulated annealing algorithm to the status´s adjusting of DHNN. This method improved the tracking performance and shortened convergence time of the network. From the Monte Carlo simulation experiments, the association performance and compute velocity of the proposed method was proved higher.