Research On CFAR Detection Method Of Railway Perimeter Alarm System Under Clutter Background

The environment around the railway is not ideal,full of many unpredictable factors,including noise,clutter and so on.How to obtain the desired information from the unknown environment is a key topic worthy of discussion and research.In view of this problem,the CFAR technology on the radar device is undoubtedly a good solution.After years of practice and optimization,this detection technology has become very mature,but many CFAR detection devices are difficult to maintain a stable operation state when subjected to external interference.After introducing the characteristics and contents of CFAR technology,this paper deeply discusses the technical improvement scheme of how to improve the accuracy of signal acquisition under the background of clutter.The main research contents can be divided into the following aspects:1.The working principles and performance indicators of radar signal detection and constant false alarm detection are summarized,four types of radar clutter and four types of radar clutter distribution models are analyzed,and the reason why this paper chooses the Weibull distribution model as the radar clutter distribution model is explained.Aiming at the differences and characteristics of different types of constant false alarm detectors,the detection performances of the mean and ordered statistics detectors are compared under the Weibull distribution model.The results show that although these two types of detectors have high detection probability in a specific environment,their detection performance is not stable enough to meet the requirements of railway perimeter alarm system.2.By analyzing and comparing the detection performance of the adaptive deletion algorithm and the mean algorithm when there are interfering targets,five improved detectors based on the ACMLD-CFAR and the mean algorithm are proposed.The estimated number of interference targets is used for detection probability analysis.The results show that the improved detector based on ACMLD-CFAR and mean algorithm can effectively improve the detection performance in the environment of disturbing target.3.Two improved adaptive deletion algorithms with excellent detection performance in multi-target background are combined with variable exponential algorithm,and the detection performance is analyzed in uniform background,multi-target background and clutter edge background using Weibull distribution model.The results show that the improved variable exponent constant false alarm detection algorithm is more sensitive to valid information,and can maintain a low false alarm probability in the environment with more clutter,even in the background of multi-target signals.It shows good detection performance,which is also the mainstream trend of the development of the radar constant false alarm detection technology of the railway perimeter alarm system in the future.