论文部分内容阅读
针对多无源传感器多维分配数据关联模型在构造关联代价时,未充分考虑位置估计不确定性所引入的误差问题,提出一种基于信息散度的数据关联算法.将伪量测信息的概率密度函数与真实观测数据的最大后验概率密度函数之间的差异性信息作为关联代价,并分别采用Kullback-Leibler散度和对称Kullback-Leibler散度来量化该差异.仿真分析结果表明,该算法具有良好的关联性能,其关联代价能更精准地反映数据关联的可能性程度.
Aiming at the multi-dimensional distribution data association model of multi-passive sensors, when the correlation cost is constructed, the error introduced by the uncertainty of position estimation is not fully considered and a data association algorithm based on information divergence is proposed. The probability density The difference between the maximum posterior probability density function and the real posterior probability density function of the observed data is used as the correlation cost and the difference is respectively calculated by the Kullback-Leibler divergence and the symmetric Kullback-Leibler divergence. The simulation results show that the algorithm has Good correlation performance, its associated costs can more accurately reflect the degree of possibility of data association.