LFM雷达对高超声速目标的数据关联研究

Multi-radar Data Association of Hypersonic Target by LFM Radar

基于统计判决的关联决策是当前多雷达数据关联的一个有效途径.然而,在采用该方法对高超声速目标数据关联的问题进行研究时,却会出现随着雷达测量精度的提高,不同雷达数据关联的正确关联概率反而下降这一现象.针对这一情况,首先,利用线性调频(linear frequency modulated,LFM)雷达合理构建了基于统计判决的高超声速目标数据关联模型;接着,在现有的统计判决机制下,进一步分析了目标高超声速运动对多LFM雷达数据关联的影响,并得出了其关联判决函数(服从非中心卡方分布)与现有的关联判决门限(服从卡方分布)相矛盾的结论;最后,结合统计学的原理,通过对目标非中心统计参量的有效求解,合理解释了上述奇怪现象的出现.仿真结果表明,在目标高超声速运动的条件下,上述奇怪现象的出现是合理的.

As a typical multi-radar data association metrics, the association method,which is based on the chi-square statistic metrics, is an effective way used for multi-radar data association in the current information fusion fields. It is found that,however, the correct probability of multi-sensor data association decreases as the radar measurement error decreases,when the target moves in a hypersonic way. In view of the strange phenomenon, the data association model,which is based on the chi-square statistic metrics, is established firstly by the linear frequency modulated radar. Then the influence of target hypersonic movement on multi-sensor data association is further discussed, and the conclusion is that the associated judgment function（ subjected to the non central chi square distribution） is contradicted against the existing associated judgment threshold（ subjected to the chi square distribution）. Finally, the strange phenomenon above,which is proposed in this paper, is explained by using the statistic theories, and an important conclusion is correspondingly drew that the existing data association approaches,which are based on the chi-square statistic metrics,have a great disadvantage on multi-sensor data association of hypersonic target. Simulation results show that, the happening of the strange phenomenon above is reasonable,when the target moves in a hypersonic way.