Research On Radar Signal Sorting Technology Of Multi-UAV Cooperative Reconnaissance

UAV(Unmanned Aerial Vehicle,UAV)has the advantages of good concealment,strong mobility,and wide sensing range,so it is widely used as the carrier of aerial information platform.Because of the limited performance of a single reconnaissance plane,it has become a vital reconnaissance mode in the modern battlefield to carry out reconnaissance missions with multiple reconnaissance planes.Radar emitter signal sorting is one of the key technologies in the modern electronic reconnaissance field,and it is also an important basic premise.The sorting effect will directly affect the operational effectiveness of the whole electronic reconnaissance system.With the extensive application of various new radar systems,the radar signals intercepted by receivers are diversified,complex,and integrated,bringing new challenges to signal sorting technology.Therefore,under the background of multi-receiver cooperative reconnaissance,from the perspective of the multi-stations time difference of arrival(TDOA)sorting,the research on how to complete radar emitter signal sorting efficiently and accurately in a complex environment is conducted.The main research contents and contributions are as follows:(1)To address the problems that the existing TDOA sorting method has a reduced correct sorting rate and low pulse utilization rate when the pulse pairing fails,a radar radiation source signal sorting method based on the three-station time difference pair sequence is proposed.Firstly,searching the same pulse emitted from the same radar radiation source in the pulse trains of two sub-stations according to the invariance of pulse characteristic parameters,and the TDOA information between the paired pulses is extracted to form the time difference pair sequence.Then,the time difference pair sequences formed by pulses from the same emitter are classified into the same set.When the pulse pairing fails to form the completed time difference pair sequence,auxiliary sorting is carried out by combining the DOA information.For very few mismatched pulses,the correlation between them and the radar emitter that has been sorted is established by the grey relation analysis(GRA)method,which provides a basis for the subsequent decision and judgment.The simulation results show that the proposed algorithm has a high sorting accuracy for a variety of repetitive frequency signals.When the loss rate and interference rate of intercepted pulses are both 15%,the sorting accuracy reaches 96.76%,which has a great improvement compared with other algorithms.At the same time,the overall running time of the algorithm is also lower than that of similar algorithms because the algorithm adopts Euclidean distance clustering in the classification process.(2)The existing TDOA histogram algorithm is easily affected by noise,and time measurement errors can cause histogram splitting.Aiming at the problems,we improved the current algorithm and combined it with the DBSCAN clustering algorithm to complete the signal sorting.Firstly,the false pulse pair in the histogram is eliminated by using the characteristic that the signal from the same emitter has little change in the direction of arrival in a short time to reduce the impact of noise.In order to solve the problem that the TDOA sets with errors are accumulated into adjacent squares,which leads to the splitting of the histogram,the "augmented" emitters generated by the splitting of the histogram are merged by comparing the distance between the center of the pulse set.Since the algorithm based on pulse pairing could not handle the mismatched pulse,the sample data was clustered and sorted by combining the DBSCAN algorithm.The simulation results show that the proposed algorithm can effectively eliminate the noise time difference in the histogram,reduce the false alarm rate,and prevent the occurrence of increasing-batch phenomenon.Meanwhile,compared with the partition clustering algorithm,DBSCAN clustering can effectively eliminate noise points and has a better effect on the clustering of radar parameters with obvious clustering characteristics.