一种新的鲁棒恒虚警检测器

New robust CFAR detector

本文基于自动删除单元平均(ACCA)方法和单元平均(CA)方法,提出了一种新的恒虚警检测器(ACGCA-CFAR)以提高CFAR检测的抗干扰性能。它的前沿和后沿滑窗分别采用ACCA和CA方法产生两个局部估计,然后取最大值作为背景噪声功率水平估计。在SwerlingII型目标假设下,推导出ACGCA在均匀背景下虚警概率Pfa的解析表达式,并与现有方案进行了比较,仿真和试验数据处理结果表明:ACGCA-CFAR在均匀背景和非均匀背景下均具有相当好的检测性能,而它的样本排序时间只有OS和ACCA的1/4。

In order to improve the interference immunity of the detector, a new CFAR detector （ACGCA-CFAR） based on automatic censoring cell averaging （ACCA） and cell averaging （CA） is presented in this paper. It takes the greatest value of ACCA and CA local estimations as the noise power estimation. Under Swerling Ⅱ assumption, the analytic expressions of Pfa in homogeneous background are derived. Comparing with other detectors, the ACGCACFAR detector has fairly well detection performance both in homogeneous background and nonhomogeneous background, while the sample sorting time of ACGCA is only a quarter of that of OS and ACCA.