航迹拓扑序列的SVD相关匹配算法  

SVD correlating algorithm for track topology sequence

作　　者：吴泽民[1] 蒋叶金[2] 任姝婕[3] 

机构地区：[1]解放军理工大学通信工程学院 [2]中国人民解放军73683部队 [3]解放军理工大学理学院

出　　处：《系统工程与电子技术》2011年第8期1881-1884,共4页Systems Engineering and Electronics

基　　金：中国博士后科学基金资助课题

摘　　要：基本拓扑序列法为适应雷达系统误差需要步进旋转拓扑序列,造成过大的计算量。通过推导系统误差与拓扑序列的近似线性关系,得到了一种基于奇异值分解(singularity value decomposition,SVD)的修正匹配算法,可以一次性地计算出匹配情况下的序列旋转角度,不但极大地提高了计算效率,而且避免了基本拓扑法中选择角度旋转步长的难题。通过仿真测试,SVD算法能减少计算时间90%以上,而且在相同系统误差情况下的正确相关概率和误相关概率性能指标都有较大的提高。In order to tolerate radar system errors,the basic topology sequence algorithm rotates topology sequence step by step,which leads to too much calculation burden.A novel singularity value decomposition（SVD） based modified topology sequence matching algorithm is introduced after inferring the approximate linear relation between system errors and topology sequence.Through direct calculation of sequence rotation angle,the modified algorithm not only greatly promotes calculation efficiency,but also avoids the dilemma of selecting angle step size.Through simulation,the SVD algorithm proves reducing calculation time more than 90%.The performance indexes of both correct correlation ratio and false correlation ratio under the same system errors are also promoted greatly.

关 键 词：数据融合 航迹相关 拓扑序列 奇异值分解 

分 类 号：TN957[电子电信—信号与信息处理]