Detection and Parameter Estimation of Strong and Weak LFM Signals in the Fractional Fourier Domain
XU Hui-fa, Feng Liu · Signal Processing · 2011
The fractional Fourier transform(FRFT)is very suitable to process the linear frequency-modulated(LFM)signals because of its unique properties.Especially,as a linear transform,FRFT can avoid the cross-terms interference in multi-component LFM signals processing.However,if we use elimination one by one method to detect the multi-component LFM signals,we will compute the FRFT of signals withα∈[0,π]for every LFM signal's detection,and search the maximum.So the computation cost is very high.In order to improve the detection efficiency of the FRFT for multi-component LFM signals,a novel detection method is presented for the strong and weak LFM signals in the fractional Fourier domain.First,introduce the detection and parameter estimation theories of multi-component LFM signals based on the elimination one by one method and clustering analysis method respectively,and analyze their advantages and disadvantages.And the novel detection method is presented combing elimination one by one method with clustering analysis method,use plane to cut the peaks of the multi-component LFM signals in the(u,α)plane,and a clustering algorithm named broad first search neighbors(BFSN)is introduced to detect the peaks.So the peaks of the LFM signals with approximative energy can be detected simultaneously. Next,use elimination one by one method to eliminate the strong signals detected.Repeat the above process until all the LFM signals have been detected.So the novel method improves the detection efficiency and the parameter estimation precision of the FRFT for stronger signals and it also eliminate the shading effects of strong signals to weak signals.The plane cutting method reduces the input data-set pointers of the BFSN clustering algorithm,and it reduces the computation cost of the novel method.Finally,simulations results verify the effectiveness of the method.