Research On Multi-frame Joint Weak Target Detection Technology Of Phased Array Radar Based On Data Correlation

In the future,there will be a large number of stealth UAVs,missiles and other weak targets in the battlefield.The airborne radar needs the ability to detect weak targets in a long distance,so as to obtain enough early warning time to carry out other evasion,attack,investigation and other tasks.But at present,with the development of target stealth technology and anti-reconnaissance technology,airborne radar is facing a severe challenge for the detection of long-distance weak targets.Multi frame detection(MFD)can effectively increase the detection distance of the target,also has higher ability to detect weak target at the same distance.However,multi frame processing needs lots of radar system resources.With the increasing requirements of multi-target detection and multi task of airborne fire control radar,especially airborne fire control radar,it is unreasonable to take multi frame joint processing in any working mode and detecting scene.Therefore,how to use the power resources,time resources and storage resources of airborne fire control radar more efficiently becomes an urgent problem.In this thesis,the optimization of multi frame joint processing algorithm and the call strategy of multi frame joint processing in the process of target tracking are studied.The main work includes:(1)Establishing the detection scene of airborne radar for airspace target.The target is divided into three regions: far,middle and near.A series of target motion scenes including uniform speed change,meandering turn,overload turn and uniform speed circle are established,Then analyzing the RCS which is changed by the change of the target attitude angle during the turning model of target?(2)The maximum posterior probability association based on dynamic programming(DP-MAP)is used in traditional multi frame joint detection,Based on the DP-MAP algorithm,this thesis considers the change of the state information of moving target between frames,then constructs a penalty function term based on the data association between frames,and a new DP-MAP algorithm based on the penalty function(DP-PMAP)is formed,then analyzing the detection performance of this algorithm,this algorithm can improve the detection probability of airborne radar detection in weak targets with the same frame number,.(3)In order to solve the problem that single frame low threshold decision preprocessing makes weak target information lose,An algorithm of DP PMAP based on data complementation is proposed(DS-DP-PMAP).Then analyze the detection performance and the data calculation and data storage of this algorithm,also compare it with the multi frame joint processing algorithm without single frame low threshold preprocessing.The analysis shows that the latter algorithms can greatly reduce the calculation and storage with the similar detection performance,.(4)A strategy which adjusts the number of accumulated frames and interval of follow-up based on track SNR is designed.in the process of target tracking.Through the point status information and the signal-to-noise ratio of the current track segment,the status information and the signal-to-noise ratio of the track segment in the next follow-up visit are predicted,Based on this,refer to the detection demand and the detection performance of DS-DP-PMAP algorithm,the next return visit accumulation frame number is set,and the return visit interval is appropriately increased/decreased based on the current return visit interval.Taking this strategy in different target motion scenes to detect and track,also analyzing the occupancy rate of system resources during the tracking process.