改进的D-S证据理论在战场目标识别中的应用∗

摘要

战场目标识别中多传感器信息融合面临各种不确定冲突信息。为此，提出了一种基于权重的证据调整方法，给予证据分配不同的权重，重新修正基本概率分配函数，利用Dempster组合规则实现信息融合。通过数值验证并与其他的改进算法对比，结果更为理想，能有效的减小冲突证据的不良影响，同时具有较高的收敛速度，降低了决策风险，仿真结果表明，算法有效、可行。%In view of uncertain conflicting information confronted by multi-sensor information fusion in radar target recognition, a new weighted D-S combination rule was put forward, then the BPA was reallocated and Dumpster’ s rule was used to achieve information fusion. The more expected result data shows that, compared with other methods, this new algorithm can reduce harmful influence of conflict evi-dence effectively, at the same time, it has higher convergence rate and reduces decision risk. Our experiment shows that the solution is ef-fective and feasible.