A Study On Radar Weak Target Detection Algorithm

How to improve the detection ability of radar to dim targets is one of the main problems in radar detection field,so it is very important to study the method of radar dim target detection.For the detection of radar weak targets,the long time accumulation algorithm is used for coherent accumulation and detection,and the pre tracking algorithm is used to improve the signal to noise ratio by non coherent accumulation.In this paper,we study these two methods respectively,and propose a nonparametric CFAR Detector in the end.The main work is as follows:First of all,according to the process of long time accumulation of range migration and Doppler migration phenomenon,through the analysis of the target echo,explains the echo signal of range migration and Doppler migration phenomenon from the angle of theory,and in the traditional uniform echo signal processing method of failure situation,put forward a self correlation function method,the autocorrelation function method to calculate the acceleration and jerk of high-speed targets,and then to compensate,finally realize the accumulation of good,also given a broadband signal multiple narrowband signals with narrowband processing algorithm,reduces the resolution,so that the target does not produce about;again,for the problem of detecting and tracking basic the function of radar,aiming at the shortage of the traditional tracking algorithm to detect,without changing the existing radar system only through Based on changing the traditional signal processing methods on the research of the detection and tracking algorithm of dynamic programming based on this algorithm,the multi frame combined data processing so as to achieve good results;finally,the constant false alarm problem,based on the traditional background clutter estimation parameter CFAR detector to meter the background clutter mismatch problem,proposes a non parametric kernel density estimation based on CFAR algorithm,this method uses kernel density function to replace the traditional clutter probability density function estimation,given the probability of false alarm and detection threshold obtained by the relationship between the probability density function and the detection threshold,thus is a very good solution by background clutter loss detection performance with the sharp decline of the.Simulation analysis shows that the proposed kernel density estimation algorithm,the autocorrelation algorithm and the dynamic programming algorithm can achieve the corresponding detection results very well.