Multi-feature fusion for target recognition based on improved D-S evidence iterative discount method

Abstract

A new multi-feature fusion method is proposed for the radar target recognition based on D-S evidence iterative discount method. Firstly, the discount factor is defined based on the multi-feature confusion matrix and basic probability assignment (BPA) function. Then, when the conflict is high, the evidence is discounted using the discount factor, and basic probability assignment function, discount factor and conflict coefficient are updated; repeat the above discounts procedure and stop the evidence source correction when the evidence conflict coefficient is less than the threshold. Finally, fusion recognition is achieved by using the revised evidence. Compared with the other fusion recognition algorithm, the simulation results show that this proposed algorithm performs better.