Research On Passive Distributed Radar Target Detection Method Based On Approximate Message Passing

As a variety of operational concepts are proposed,new types of weapons are being updated faster and faster,radar as the eyes of the battlefield also needs continuous innovation and development to meet the challenges.Among them,passive distributed radar not only has natural advantages in radar survival,but also compared with single radar base station detection,distributed radar multi-base station fusion detection has higher time and space coverage and better ability to suppress target flicker.At the same time,passive distributed radar can make full use of existing electromagnetic resources to form a networked,low-cost regional air defense early warning system,which has far-reaching significance in the entire national defense construction.However,the proposed passive distributed radar target detection algorithms are usually obtained on the basis of directly using the correlation between the signals received by different radar base stations,but do not make good use of the received signals that are approximately sparse or in the frequency domain.So this paper carries out the research of passive distributed radar centralized and distributed target detection algorithms based on the feature.At the same time,in order to solve the problem of sparse recovery of large-scale data in detection,it is considered to bring approximate message passing algorithms into it.The corresponding target detection algorithms CAMP-GLRT and DAMP-GLRT are derived and analyzed.The specific expression is as follows:1.The passive distributed radar system signal model under the sparse representation of the received signal is established,and the basic target detection algorithm is derived using the generalized likelihood ratio detection method.Different from the traditional sampling method,the model of the received signal under sparse representation can use a sampling rate far lower than 2 times the highest signal frequency for compressibility processing.At the same time,when the generalized likelihood ratio detection method is used to derive the target detection algorithm,the paper presents two different centralized target detectors for the condition of whether the perception matrix of each radar base station is the same.After simulation analysis,it is found that only when the perception matrix of each radar base station is the same,the algorithm using the generalized likelihood ratio detection method can detect the target.2.The centralized target detection processing algorithm CAMP-GLRT based on the approximate message passing algorithm is studied.In order to solve the problem that the generalized likelihood ratio detection method cannot be used for target detection when the perception matrix of each radar base station is different,this paper proposes a centralized target detection processing algorithm CAMP-GLRT.The algorithm also first transmits the signals of each radar base station to the fusion center,but the fusion center adds a process of sparse recovery of the received signal using the approximate message passing algorithm based on the generalized likelihood ratio detection method as the target test criterion.Derive the corresponding test statistics.The simulation results show that the CAMP-GLRT detector can not only maintain stability and effectiveness under the condition that the perception matrix of each radar base station is different from each other,but also the number of iterations required for the convergence of the CAMP-GLRT algorithm is small,and the calculation of each iteration The complexity is low,so that the time to make a detection decision is very small.3.The distributed target detection processing algorithm DAMP-GLRT,which combines the advantages of the approximate message passing algorithm and decentralization,and its improved target detection algorithm are studied.In the DAMP-GLRT algorithm,each radar base station performs local processing on the collected data,and then obtains the global processing result through the communication protocol between the base stations,and then performs the next iterative operation until convergence,and finally each node makes the same decision processing result.DAMP-GLRT does not require a fusion center to collect signals from all radar base stations,and maintains effective target detection capabilities while achieving "decentralization" capabilities.However,the DAMP-GLRT algorithm handles a huge amount of traffic globally,so an improved DAMP-GLRT algorithm is proposed.Although the algorithm loses little detection performance,it can greatly reduce the amount of communication.