Multi-Sensor Target Recognition Based-on Multi-Period Improved DS Evidence Fusion Method

摘要

This paper presents a novel multi-period improved DS evidence fusion method to addresses the problem of target recognition caused by information uncertainty and evidence combination in complex battlefield environment. A fusion structure consisting of four levels and two aspects is used in the proposed method to improve the DS evidence processing performance. Additionally, historical recognition information is fully utilized in the proposed method through multiple cycles of information fusion to complete the final recognition. Numerical simulation results confirmed the optimality of the proposed method and demonstrate it can effectively solve the problem of false recognition caused by the target density and the response deception while improving the identification performance.