论文部分内容阅读
本文是作者在论文[1～3]研究基础上的继续和深入,作了理论上的结论并给出了结果。这箱论文的目的是多方面的。首先,对数学背景作了简要的修正,以便推导适于任何类型的目标和杂波概率密度和自相关函数的雷达检测算法。应用这个理论的相应条件是要导出目标和杂波过程的适当的数学模型。我们特别对于对数正态分布和韦布尔(Weibull)分布的情况下作了推导。并且在对相干回波序列的模型进行推导时,得到了非常好的结果。然后,引入“漂白—高斯化”滤波器的重要概念,并以其为基础,得到了雷达检测方案。当杂波的幅度概率密度呈对数正态分布或韦布尔分布时,应用上述理论得到了完全新的检测方案,并计算了相应的检测性能。本文另一新颖之处是所提出的检测方案具有自适应特点。详细地说,提出了“漂白—高斯化”滤波器的权值的直接估值方法。本文给出了检测失损与距离单元数目之关系,权值的平均值是沿这些距离单元进行估计的。本文对不同的被处理的脉冲数再和不同的杂波和目标信号的参数时的检测损失作了计算。另外,研究了直接计算恒虚警率检测门限的自适应特性,并估算了相应的检测损失。
This article is the author on the basis of the paper [1 ~ 3] on the basis of the continuation and in-depth, made a theoretical conclusion and gives the results. The purpose of this box is multifaceted. First, the mathematical background is briefly modified to derive a radar detection algorithm suitable for any type of target and clutter probability density and autocorrelation functions. The corresponding condition for applying this theory is to derive the appropriate mathematical model of the target and clutter processes. In particular, we derive the logarithmic normal distribution and the Weibull distribution. And when deriving the model of the coherent echo sequence, very good results have been obtained. Then, the important concept of “bleaching-Gaussian” filter is introduced and based on which, the radar detection scheme is obtained. When the clutter amplitude probability density logarithmic normal distribution or Weibull distribution, the application of the above theory has been completely new detection scheme, and calculate the corresponding detection performance. Another novelty of this paper is that the proposed detection scheme has adaptive characteristics. In detail, a direct estimation method of the weight of the “bleach-Gaussian” filter is proposed. In this paper, the relationship between the detection loss and the number of distance units is given. The average of the weights is estimated along these distance units. In this paper, we calculate the detection loss of different processed pulse numbers and parameters of different clutter and target signals. In addition, we study the adaptive characteristics of directly calculating the threshold of CFAR detection and estimate the corresponding detection loss.