%0 Journal Article
%T 多个不可分辨目标群的联合检测与估计误差界<br>Error Bounds of Joint Detection and Estimation for Multiple Unresolved Target？？Groups
%A 连峰
%A 王婷婷
%A 韩崇昭
%J 西安交通大学学报
%D 2015
%R 10.7652/xjtuxb201511015
%X 针对杂波和漏检同时存在时多个不可分辨目标群联合检测与估计的性能评价问题,在随机有限集框架下,利用信息不等式和最优子模式分配距离提出了该问题的误差(下)界。首先将多个不可分辨目标群的状态建模为多Bernoulli随机有限集,并利用Mahler提出的连续个体目标数假设建模群目标测量似然函数;然后,结合最大后验概率检测和无偏估计准则获得了建议的误差界。仿真实验展示了该误差界随杂波密度和传感器检测概率的变化趋势,并利用不可分辨目标群势概率假设密度滤波器和不可分辨目标群势平衡多目标多Bernoulli滤波器对该误差界的有效性进行了验证。实验表明,利用该误差界可以对现有的不可分辨目标群联合检测与估计算法的性能进行有效衡量,使其在不同杂波密度和检测概率下的平均相对误差不超过7%。<br>Aiming at the performance evaluation of joint detection and estimation (JDE) for multiple unresolved target？？groups in the presence of clutters and missed detections, an error (lower) bound is proposed using information inequality and optimal sub？？pattern assignment distance within the random finite set (RFS) framework. Firstly, the states of multiple unresolved target？？groups are modeled as a multi？？Bernoulli RFS, and the group observation likelihood is modeled based on the assumption of the continuous individual target number proposed by Mahler. Then, the maximum posterior probability detection criteria and unbiased estimation criteria are used in deriving the proposed bound. Experimental results show that the proposed bound can effectively indicate the performance limits of JDE algorithms for the existing unresolved target？？groups, and the average relative error is less than 7% for various clutter density and detection probability
%K 误差界
%K 不可分辨目标群
%K 联合检测与估计
%K 随机有限集&lt
%K br&gt
%K error bound
%K unresolved target？？group
%K joint detection and estimation
%K random finite set
%U http://zkxb.xjtu.edu.cn/oa/DArticle.aspx?type=view&id=201511015