一种长基线被动传感器数据关联方法

A long baseline passive sensor data association method

为了提高测向交叉定位系统多目标环境下的去鬼点性能,提出了一种长基线被动传感器数据关联方法.首先利用被动传感器量测的视线矢量信息进行交叉定位并计算定位误差协方差矩阵,然后基于两个定位点构造了一个卡方分布的检验统计量,将多目标条件下的去鬼点问题转换为假设检验问题.为了分析该方法的性能,定义了真实目标的正确关联概率和鬼点目标的错误关联概率,并在使用两部和三部被动雷达的情况下分别与倾斜角方法进行了仿真对比.仿真结果表明所提出的方法在使用三部被动雷达或多目标与被动雷达共面时关联性能更好.

This paper proposes a long baseline passive sensor data association method to improve the de-ghosting performance in the multi-target environment of direction-finding cross positioning system.Firstly,the line of sight(LOS)vector information measured by passive sensors is used for cross-positioning and the covariance matrix of positioning error is calculated.Then a Chi-square distribution test statistic is constructed based on the two positioning points,and the de-ghosting issue in multiple target conditions is transformed into a hypothesis testing issue.Finally,in order to analyze the performance of the proposed method,the correct association probability of the real target and the false association probability of the ghost target are defined,and the simulation comparison is also made with the hinge angle method in the case of using two and three sensors,respectively.The simulation results show that the proposed method has better correlation performance when three sensors are used or multiple targets are coplanar with sensors.