quotation:		[Copy]
		[Copy]

This Paper:Browse 3088 Download 2027
一种网格概率多目标数据关联算法
周航,冯新喜
0 Fontlarge +\|Default\|Small
(空军工程大学电讯工程学院，西安 710077)

摘要:

针对多目标跟踪中数据关联问题，提出了一种基于网格概率的数据关联算法。该算法首先计算出关联门内全部回波的关联系数，然后利用网格概率的思想将关联系数小的有效回波的系数值配权给其它有效回波，同时对于两跟踪波门交叉区域内的有效回波也使用配权法分别分配给两个跟踪波门内的其它邻近回波，最后通过概率加权进行数据关联。仿真结果表明，该算法能够有效解决多目标跟踪中的数据关联问题，并且有较高的关联正确率，而且 CPU占用时间较短。

关键词: 多目标跟踪网格概率回波关联系数有效回波

DOI：

基金项目:

A multi-target data association algorithm based on grid probability

ZHOU Hang,FENG Xin-xi

(Telecommunication Engineering Institute,Air Force Engineering University,Xi′an 710077,China)

Abstract:

To solve the data association problem in multitarget tracking, a novel gorith m of data association is proposed based on grid probability. First, this algori thm calculates the whole effective echoes correlative coefficients, then uses gr id probability concept to allot the value of effective echoes with smaller cor relative coefficient to other effective echoes.For the effective echoes in interaction area of tracking wavedoor, it uses the same method to assign the v alues to the other adjacent echoes within the two tracking wavedoors respectiv ely.Finally the correlative coefficients of the echoes are correlated through pr obability-weighting. Simulation results show the algorithm can effectively solve the problem in multitarget tracking, and has a higher accurate rate of associ ation. Besides,the CPU occupies a relatively short time.

Key words: multi target tracking grid probability correlative coefficients effective ec hoes