摘要
文中基于删除平均(CM)方法和单元平均(CA)方法,提出了一种新的恒虚警检测器(CMCAGO-CFAR)。它采用CM和CA方法产生两个局部估计,再取二者的最大值作为背景噪声功率水平估计。在SwerlingII型目标假设下,文中推导出CMCAGO-CFAR在均匀背景下虚警概率Pfa和检测概率Pd及多目标环境下检测概率Pd的解析表达式,并与其它方案作了比较。分析结果表明CMCAGO的优势主要体现在非均匀背景中,它在杂波边缘的虚警控制能力明显优于OS,对多目标干扰也呈现了较好的鲁棒性,它以均匀背景中较小的代价换取在多目标干扰情况下检测性能的较大提高。
A new constant false alarm rate （CFAR） detector （CMCAGO-CFAR） based on censored mean and cell averaging is presented in this paper . It takes the greatest value of censored mean （CM） and cell averaging （CA） local estimation as a noise power estimation. Under SwerlingⅡ assumption, the analytic expressions of Pfa and Pa in homogeneous background are derived, and the analytic expression of Pa in multiple target situations is also derived. In contrast to other detectors, the results show that the advantage of CMCAGO lies in nonhomogeneous background, the ability of CMCAGO to control rise of false alarm is more effective than that of OS, while CMCAGO also exhibits good robustness in multiple target situations. A compromise can be made in homogeneous background leading to a great improvement on the performance of CMCAGO in multiple target situations.
出处
《弹箭与制导学报》
CSCD
北大核心
2007年第2期309-312,共4页
Journal of Projectiles,Rockets,Missiles and Guidance
基金
国防预研基金资助
关键词
检测
恒虚警
删除平均
单元平均
detection
constant false alarm rate
censored mean
cell averaging
