改进的DSmT算法及其在C^4ISR系统中的应用  

Improved Method Based on DSmT and Its Application in C^4ISR System

作　　者：覃东升[1,2] 苗壮[3] 王勇[4] 

机构地区：[1]西北工业大学航天学院 [2]中国人民解放军91550部队 [3]中国人民解放军63928部队 [4]中国西南电子技术研究所

出　　处：《电子科技大学学报》2014年第4期591-595,共5页Journal of University of Electronic Science and Technology of China

基　　金：航天技术支撑基金(2013-HT-XGD-014)

摘　　要：提出了一种基于DSmT的快速mass函数收敛算法。在工程应用中,由于矛盾焦元在决策时没有实际意义,所以在证据融合时,只计算由辨识框架元素形成的焦元与未知焦元的基本置信指派,而不需要计算矛盾焦元的基本置信指派。这样能够有效地降低计算复杂度,并提高mass函数的收敛速度。将该算法作为雷达辐射源识别的方法应用到了C4ISR系统中,仿真表明,该方法比传统方法在高冲突情况下可以更准确、更快速地识别雷达目标。Dezert-Smaradache thoery （DSmT） can well deal with the fusion of conflicting information, but the number of focal elements will greatly increase with the increment of the number of elements in frame of discernment, which causes the big computation burden. An improved method based on DSmT is proposed in this paper. In engineering applications, the conflicting focal elements are not very useful in the decision-making. Thus, we just calculate the basic belief assignment （bba） of the focal elements included in the frame of discernment, and the bba of conflicting focal elements is ignored. This is able to efficiently reduce the computation complexity, and improve the speed of convergence of mass function. The new proposed method is applied in the C4ISR system for identifying the radar radiation source. The simulation results show that new method can more accurately and quickly recognize the radar target with conflicting information than classical method.

关 键 词：C4ISR系统 DSmT算法 MASS函数 雷达辐射源识别 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]