Radar Signal Recognition And Parameter Estimation Based On FRFT

Today's world situation is changing rapidly.Modern military strength affects the outcome of the game between all parties,and therefore directly determines the development direction of the field of electronic countermeasure technology.The electromagnetic environment of the battlefield is becoming more and more complex,and traditional radar signal analysis methods are gradually difficult to effectively complete tasks,and the challenges are imminent.The fractional Fourier transform as a generalized form of the traditional Fourier transform can make the signal to rotate arbitrarily in the time-frequency plane to obtain extensive signal properties which is suitable for analyzing the characteristics of the non-stationary signals.This paper focuses on radar signal recognition and parameter estimation methods based on FRFT and summarizes the work as follows:Firstly,the fundamental theory of FRFT is studied,aiming at the problem of recognizing typical modulation signals,a typical modulation recognition method based on FRFT is proposed.The characteristics of the envelope curve of the FRFT spectrum of the signal and the instantaneous autocorrelation of the signal are combined with the idea of a classification tree,realized the recognition of four kinds of single modulation signals.Aiming at the problem of recognizing hybrid modulation signals,a hybrid modulation recognition method based on FRFT and phase jump detection is proposed.the algorithm uses the characteristics of the envelope curve of the FRFT spectrum of the signal square for coarse classification,and then uses the phase characteristics for fine classification,the recognition of five modulation signals is realized from coarse to fine.Simulation results show that the algorithm still has a good recognition effect under low SNR.Then,for single-component LFM signals,in view of the computational redundancy caused by global search in parameter estimation based on FRFT,a FRFT extremum search method based on continuous colony optimization is proposed,it obtains the coarse estimated FRFT order through the two-dimensional coarse search,so as to limit the search interval and bring it into the CACO algorithm as a prior value for optimization to achieve the purpose of improving the search efficiency.The simulation results show that the search efficiency of the algorithm is improved,and the accuracy of the parameter estimation can be guaranteed at the same time.Finally,for multi-component LFM signals,aiming at the problem of repeated search and estimation accuracy in parameter estimation based on FRFT,a parameter estimation method of multi-component LFM signal based on Lv's Distribution and FRFT high-resolution is proposed.In this algorithm,component detection and coarse parameter estimation are implemented by LVD,the coarse estimation value is used as the prior value to limit the FRFT search interval to achieve the purpose of reducing the calculation times.and improving the accuracy of parameter estimation by refining the FRFT spectrum with Zoom-FRFT,simulation results verify the effectiveness of the new algorithm.Aiming at the problem of acquiring local characteristics of signals,the basic principles and implementation methods of Short Time Fractional Fourier Transform are analyzed in detail,and a multi-component LFM signal detection and processing method based on STFRFT is introduced,the algorithm combine with the idea of CLEAN to process signal components step by step,and achieves good estimation results.