基于多传感器信息融合的异步点迹关联方法  

ASYNCHRONOUS DOTTED CORRELATION BASED ON MULTI-SENSOR INFORMATION FUSION

作　　者：刘静涛 路娜 周君会 Liu Jingtao;Lu Na;Zhou Junhui(91977 Unit,Beijing 100036,China;School of Telecommunications Engineering,Xidian University,Xi’an 710071,Shaanxi,China)

机构地区：[1]91977部队,北京100036 [2]西安电子科技大学通信工程学院,陕西西安710071

出　　处：《计算机应用与软件》2022年第1期296-300,共5页Computer Applications and Software

摘　　要：对于异步场景下的数据关联,通过时间配准将异步点迹集合同步化的方法计算复杂且误差较大,严重影响关联效果。在不进行时间配准的前提下,提出一种基于集合经验模态分解(Ensemble Empirical Mode Decomposition,EEMD)来提取趋势项的异步点迹关联方法。仿真结果表明,该方法整体上的关联效果好于基于时间配准后的KNN算法。在目标数量适中的场景下有较好的关联效果。For the data association in the asynchronous scenario,the asynchronous point trace set is synchronized through time registration.The calculation complexity of this method is complex and the error is large,which seriously affects the correlation effect.Without time registration,this paper proposes a synchronization dotted correlation method based on ensemble empirical mode decomposition(EEMD)to extract trend items.The simulation results show that the overall correlation effect of the proposed method is better than KNN algorithm based on time registration.There is a better correlation effect in a scenario with a moderate number of targets.

关 键 词：信息融合 点迹关联 异步点迹关联 

分 类 号：TP301[自动化与计算机技术—计算机系统结构]