Study On Intelligent Sorting And Working Pattern Recognition Method Of AESA And SAR Radar Signals

AESA(Active Electronically Scanned Array)radar has many excellent characteristics such as variable characteristics of the transmitted signal,beam agility,and low signal-to-noise ratio.At present,the new multifunctional radar arranges some resources to complete the imaging function while performing conventional search and track work patterns.Traditional electronic reconnaissance methods are difficult to classify and identify such signals.In order to meet the radar reconnaissance challenges brought about by AESA and SAR radar signal sorting and working pattern recognition,this paper carries out the research on intelligent sorting and working pattern recognition methods of AESA and SAR radar signals.This paper analyzes the characteristics and signal parameters of AESA and SAR radar.On this basis,it proposes an adaptive parameter sorting method based on grid and density,and from the sparseness of high-dimensional features of radar signals,identify working mode of single-function radar signals based on high-dimensional subspace clustering algorithms,and verify the effectiveness of the algorithm.The main research contents and achievements are as follows:1.The signal parameters of the multi-function radar are studied against waveform agility of the signal transmitted by the multifunctional radar.The signal parameters are extended on the basis of the conventional radar signal parameters and modeled based on the joint parameters of the signals of multiple working modes.This provides the basis for subsequent signal sorting and work pattern recognition.2.Against the problem that the traditional radar signal sorting algorithm needs prior information to set the threshold,a self-adaptive sorting method based on grid density parameters is studied.Firstly,the influence of the traditional grid quantization scales on the clustering results is analyzed in depth,and a parameter adaptive processing method is proposed.The dynamic grid partitioning method and the mobile grid technology are jointly used in the algorithm to effectively avoid creating unnecessary empty grids in the clustering process.After that,the grid merging and boundary processing methods that take into account the transition meshes are studied.Finally,based on the above algorithm,the AESA radar and SAR radar signals are sorted and simulated,and the sorting effect in the case of pulse loss and noise interference is analyzed.The effectiveness of the algorithm is evaluated from the three perspectives of correct sorting,missed sorting,and wrong sorting,which proves the effectiveness of the algorithm in intelligent sorting of multi-functional radar signals.3.According to the principle of high-dimensional feature sparseness of multifunction radar signals,a work pattern recognition algorithm based on high-dimensional subspace clustering is proposed.The kernel density estimation method is introduced into the algorithm.On the basis of the above processing,data conversion and the dense subspace are performed to obtain clusters of various operating modes.Finally,under the circumstance of pulse loss and noise interference,the signal pulses of the multiple working modes of the single-function radar after sorting are used to simulate,and the recognition results are evaluated from the two aspects of recognition accuracy and recognition error.It is verified that this method is insensitive to noise data and has a good recognition effect for multiple working modes,which provide a new idea for the recognition of working modes of multi-functional radars.