Research On Multistatic Sonar Target Detection Method Based On Constant False Alarm Technology

With the continuous development of intelligence of our country’s underwater military equipment,multi-base sonar has become a hot research direction in the field of underwater target detection due to its wide range of functions and diverse functions.Based on this background,this paper studies the multistatic sonar target fusion detection method based on constant false alarm technology,and verifies the usability of the algorithm through simulation and field experiments.First,the paper studies the principle of the classic constant false alarm detection technology,respectively is the unit average constant false alarm detector(CA-CFAR),the average selection of the constant false alarm detection Detector(GO-CFAR),Average Select Small Constant False Alarm Detector(SO-CFAR),Orderly Statistics Constant False Alarm Detector(OS-CFAR)and Automatically Delete Mean Constant False Alarm Detector(ACMLD-CFAR).The detection performance of them in a uniform environment,an environment with interfering targets,and a reverberant edge environment is analyzed by simulation.Secondly,two adaptive constant false alarm detectors are studied,which are based on the variable index constant false alarm detection algorithm(VI-CFAR)and based on the improved neural network constant false alarm detection algorithm(NN-CFAR).The improved NN-CFAR is proposed on the basis of VI-CFAR.The algorithm uses neural networks to classify and recognize background environment types,and adds the ACMLD-CFAR option to the detection strategy to improve the system’s ability to cope with multi-target detection environments.ability.The neural network can be used to distinguish the background reverberation more accurately.At the same time,the advantages of the four classic CFARs are combined,and the good adaptability and detection performance of the detector are verified by simulation.Finally,in the fusion detection part,the hard decision fusion criteria such as K rank N and Neiman Pearson are first used to study and analyze the detection probability and effective detection range of the multistatic sonar fusion detection system in the case of one-shot tworeceiving and one-shot three-receiving.On this basis,a soft decision fusion criterion with the characteristics of multi-threshold structure is proposed.The detection data of each base is characterized by four characteristic quantities,and after transmission to the fusion center,the target amplitude and threshold are restored using the mean value method.Then use global statistics and global thresholds to judge the presence or absence of underwater targets.The algorithm was verified by the data processing results of the lake test.