Research On Multifunctional Radar Inter-pulse Feature Extraction And Recognition

Multi-functional radar can perform a variety of tasks by one radar,including search,tracking and imaging.It develops rapidly with the progress of the level of electronic devices,which is reflected in the enhancement of beam directivity and parameter agility.The complex characteristics of multifunctional radar make the signal interception and signal analysis become the difficulty and focus in the field of electronic reconnaissance.Starting from the existing reconnaissance data,in-depth mining of the characteristic information of multi-functional radar signal is of great significance for radar description,radar model recognition and radar countermeasure.Based on the above analysis,this paper mainly studies the following contents:Firstly,the problem of identifying PRI modulation types on continuous PRI sequences is studied.In view of the current situation that the current research focuses on dealing with single type PRI sequence and cannot recognize PRI modulation type in undivided sequence,an RP-YOLO recognition algorithm based on recurrent graph and Yolo algorithm is proposed,which transforms the problem of identifying PRI modulation type into the problem of image multi-target detection.Simulation results show that compared with the traditional sliding window method,the RP-YOLO algorithm proposed in this paper can more effectively identify and locate the sequences corresponding to different PRI modulation types under non ideal reception conditions.Secondly,the problem of repeated parameter pattern extraction is studied.In order to make full use of reconnaissance data and analyze the parameter jump law of radar target,firstly,the relationship between parameter jump law and radar word is studied,then the concept of motif mining in time series is introduced,and a fast motif mining algorithm based on grid index and KD tree structure is designed.Further,by analyzing the limitations of motif mining in the process of repeated parameter pattern extraction,motif connection and similar fragment aggregation algorithms are designed respectively,and finally the extraction of repeated parameter patterns with arbitrary length is realized,which lays a foundation for the description of parameter jump law of multi-functional radar signal.Finally,the modeling and recognition of multifunctional radar are studied.Based on the previous repeated parameter pattern extraction results,a radar modeling method based on parameter pattern combination is proposed,and based on this description method,a multifunctional radar recognition method based on key parameter pattern matching is proposed.This method uses the integrated matcher to match the key parameter patterns in the pulse flow,and realizes the recognition of radar models with strong interpretability and good effect.