现代雷达系统中,杂波统计特性往往无法预先确定,此时针对性较强的参量检测方法就失去了恒虚警能力,因此非参量方法已成为一个重要的研究方向。通过对同一距离单元不同观测时间区间上由广义符号检测器形成的多个检验统计量进行排序后选大或选小,提出了两种广义符号检测器的改进型,即基于广义符号统计量的最大或最小选择检测器。在高斯杂波和实测海杂波数据下,分析了上述两种检测器对于Swerling 0和SwerlingⅡ目标的检测性能,研究表明,两种新型检测器的检测性能均优于GS检测器。
In modern radar systems,the clutter's statistic characters are often unknown.With this clutter,parametric detectors lose the capability of CFAR.So nonparametric detectors become very important.Two improved generalized sign detector are proposed respectively by choosing the greatest and smallest test statistics which are generated by the generalized sign detector using the returns coming from the same range cell at different intervals of observation time.They are called the smallest and greatest selection detector based on a generalized sign test statistic.The Gaussian clutter,real sea clutter data are used to analyze the detection performances of the two detectors under Swerling 0 and Swerling Ⅱ target model.The results show that the detection performances of the two new detectors are better than that of the GS detector.