基于Cramer-Rao界的密集目标统计分辨限  

Statistical Resolution Limit for Closely Spaced Targets Based on Cramer-Rao Bound

作　　者：陆顶洪[1] 李阳[1] 梁超杰[1] 王欢 

机构地区：[1]北京理工大学信息与电子学院,北京100081 [2]中国兵器工业第203研究所,陕西西安710065

出　　处：《北京理工大学学报》2014年第7期742-747,共6页Transactions of Beijing Institute of Technology

基　　金：高等学校学科创新引智计划资助(B14010)

摘　　要：基于Cramer-Rao界,对Smith^([10])给出的密集信号统计分辨限(statistical resolution limit,SRL)进行了进一步研究,给出了一般结果.推导得出了两个点目标空间频率差Cramer-Rao界(CRB)的闭式表达式,并针对目标密集分布情况对此CRB进行了渐进分析,进一步由Smith准则确定了雷达系统密集目标分辨限的一般形式.SRL是目标功率比、目标回波初相差和信噪比(R_(sn))的函数.探讨了不同功率比和初相差条件下SRL与两个目标R_(sn)的关系.极大似然估计(MLE)仿真结果表明,在可分辨区内MLE的实验性能达到了CRB;在本文预测的分辨限处MLE的性能开始偏离CRB曲线.仿真结果验证了本文给出的统计分辨限是准确、合理的.An extension study of Smith＇s result was proposed.An exact Cramer-Rao bound expression of frequency seperation was deduced.The Cramer-Rao bound expression was analyzed asymptotically for the closely spaced targets case.A general close form of statistical resolution limit（SRL） was provided based on the asymptotic results.The SRL proposed is a function of power ratio,phase difference,and SNR.It is found that the power ratio and phase difference affect the relationship between SRL and SNR severely in further examination of the effect of the power ratio and phase difference.Maximum likelihood simulation results show that the variance of the MLE is close to the CRB in the resolvable area,but departs the CRB at the predicted resolution limit point.Simulation results proved that the statistical resolution limit proposed is correct and accurate.

关 键 词：统计分辨限 CRAMER-RAO界 Smith准则 密集目标 

分 类 号：TN914.4[电子电信—通信与信息系统]