论文部分内容阅读
为了更好地解决高冲突证据的融合问题,提出一种3条证据直接融合的改进D-S算法.该算法首先根据证据支持贴近度函数给出识别框架下各焦元支持度的计算方法;其次根据三维证据直接融合产生的冲突因子的性质及各焦元的支持度,提出一种基于D-S证据组合规则的冲突信息加权分配算法;最后以多传感器多目标识别系统为背景进行仿真实验.理论分析和仿真结果表明,基于三维证据直接融合的改进D-S算法具有较强的抗干扰性能,能有效融合各种冲突信息,提高目标识别概率.
In order to solve the problem of the fusion of highly conflicting evidences, a modified DS algorithm is proposed, which directly integrates the evidence.The algorithm firstly gives the calculation method of the support degree of each focal point under the framework of the recognition according to the evidence support degree function, and then, In the end, a collision information weighted allocation algorithm based on DS evidence combination rule is proposed based on the nature of the conflict factors generated by the direct integration of 3D evidence and the support of each focal point. Finally, the simulation experiment is carried out with the multi-sensor multi-target recognition system as the background. The simulation results show that the improved DS algorithm based on direct fusion of 3D evidence has strong anti-jamming performance, which can effectively fuse all kinds of conflict information and improve the target recognition probability.