针对测向交叉定位系统数据关联中存在虚假定位的问题,提出一种改进的分层快速关联算法。该算法先通过挖掘各传感器的相互位置关系与其对同一目标所形成方位量测之间的几何规律,利用候选关联的方位量测构造检验统计量,进行方位粗关联,排除了大部分虚假候选关联,减小了运算量;再对剩余候选关联进行分层组合,提取出正确的关联集合。仿真结果表明,文中算法可快速、准确的排除虚假定位点。

Focused on the problem that data association will produce false intersection in passive cross location system of multiple sensors, an improved hierarchical fast association algorithm was presented. As for the algorithm, geometrical rule between mutual position relation to sensors each other and their azimuth measures to the same target was dug out, the azimuth of the candidate associations was used to construct the statistic test and eliminate some false intersection points. Then hierarchical combination was conducted for the remaining candidate associations to pick up true associations. Simulation results show that the algorithm can fast and correctly eliminate false intersection points.