Research On Multi--mode Radar Signal Sorting Technology In Complex Electromagnetic Environment

With the widespread use of radar in modern warfare,the pulse density in the electromagnetic environment has rised sharply.In response to the electronic reconnaissance form the enemy,the workding modes of modern radar have become more and more abundant.It has become an urgent problem to find a multi-mode radar sorting algorithm that can be used in complex electromagnetic environment.In a complex electromagnetic environment,increasing batches and missing batches are the biggest challenges for radar sorting algorithms.when the modulation parameters of multiple radars are sim-ilar,existing sorting algorithm often sort them into one radar,which will result in missing batches.Besides,due to the multi-mode of modern radar,the modulation parameters usually change within a range,which will result in increasing batches.In this paper,the correlation index between radar pulses and radar radiofrequency fingerprint are used to reduce the number of increasing batches and mmissing batches in sorting results.In order to realize the sorting algorithm that can be performed in real time,this paper also studies the algorithm to cut the radar pulse stream and the algorithm of to combine the sorting results of sub stream.The work of this paper is as follows.1.The problem of sorting algorithm based on modified PRI transform is analyzed in multi-mode radar scene.The concepts of column purity and overlap rate are proposed to reduce the occurrence of additional batches and missing batches.The improved algorithm achieves 88% accuracy in the scene where 10 unknown radars coexist.2.The RF fingerprint characteristics of various pulse modulation radar signals such as BPSK,QPSK,FSK,LFM,etc.were studied.the method based on based on differential constellation trajec-tory map and the method based on the combination of raw AD data in transient phase and intra-pulse parameters in steady phase are performed to extract radar fingerprint features.3.Radar RF fingerprint feature is used to replace traditional features.By using CNN algorithm in known radar sorting and using DBSCAN algorithm in unknown radar sorting,the improved algorithm reduces the errors effectively.4.A sequential algorithm,which based on the maximum posterior probability criterion,is pro-posed to update radar parameters.By adding the posterior probability in the current observation win-dow to the posterior probability of the existing measurement results,the accuracy of sorting increases with the observation time increased.5.A radar online sorting processing framework,which combining the improved radar sorting algorithm and the radar parameter sequential updating algorithm,is designed to be used in a complex electromagnetic environment.