使用杂波先验信息和参考数据的距离扩展目标检测  

Range Spread Target Detection Using Clutter Prior Information and Reference Data

作　　者：邹鲲[1,2] 廖桂生[1] 李军[1] 刘自富 

机构地区：[1]西安电子科技大学雷达信号处理国家重点实验室,西安710071 [2]空军工程大学信息与导航学院,西安710077 [3]星海通信科技有限公司,福州350008

出　　处：《信号处理》2014年第2期141-148,共8页Journal of Signal Processing

基　　金：中国博士后科学基金(2012M521744);国家自然科学基金(61271292)

摘　　要：在雷达目标检测中,待检测单元的杂波协方差矩阵估计质量直接影响到检测性能。杂波协方差矩阵的估计有两种途径:利用一定数量的参数数据,或者是利用杂波的先验信息。本文针对距离扩展目标,研究同时利用杂波场景的先验信息与一定数量的参考数据的检测方法。基于贝叶斯方法,提出了单步广义似然比检验和双步广义似然比检验两种检测算法。与仅使用先验信息或参考数据的常规检测算法相比,该算法具有更大的灵活性。计算机仿真表明,综合利用先验信息与参考数据,可以显著提高距离扩展目标检测性能。且在小样本条件,以及先验失配条件下,本文给出的算法具有更好的稳健性。For the radar target detection,the estimation performance of clutter covariance matrix of the cell under test is directly related to the detection performance.There are two approaches to estimate the clutter covariance matrix,using some reference data,or using some prior information of the clutter.In this paper,we consider the problem of range spread target detection using both prior information of clutter covariance matrix and some reference data.One-step generalized ratio test （GLRT） and two-step GLRT are proposed based on Bayesian approach.Compared with conventional detection algorithms,which only uses prior information or reference data,these detection algorithms have more flexibility.The computer simulation results show that,using the both prior information and reference data,the detection performance can be improved significantly.And the algorithms proposed in this paper are more robust than other algorithms under the small sample size condition or mismatched prior information condition.

关 键 词：信号检测 距离扩展目标 先验信息 参考数据 

分 类 号：TN957.51[电子电信—信号与信息处理]