Research On Application Of FRFT In Detection And Filtering For LFM Signal

LFM(linear frequency modulation)signal is a typical non-stationary signal,which is widely used in radar,communication,sonar and other systems.It is significant to detect and estimate LFM signal.The LFM signal is inevitably disturbed during the transmission process,so how to separate the required LFM signal from the noise and multi-component signal,that is the filtering of LFM signal,has become an urgent subject to be studied.The finite-horizon LFM signal can present the energy aggregation phenomenon through the fractional Fourier transform(FRFT),so FRFT has become one of the effective tools for analysing LFM signal.Firstly,short-time Fourier transform(STFT),Wigner-Ville distribution(WVD)and Wigner-Hough transform(WHT)are used to detect single and multi-component LFM signals.Secondly,the theory of FRFT is studied,and the model of LFM signal detection and parameter estimation is constructed.The parameter estimation principle of single and multi-component LFM signals in fractional Fourier domain is described by mathematical derivation.The influence of dimensional normalization,sampling frequency and signal-to-noise ratio(SNR)on the performance of LFM signal parameter estimation is discussed.The detection performance of FRFT,STFT,WVD and WHT in LFM signal is analyzed and compared.The results show that FRFT is more suitable for the processing of LFM signal.Once more,the specific process of the LFM signal filtering algorithm in fractional Fourier domain is given.The models of single noisy LFM signal filtering and multi-component LFM signal separation are established.The simulation results prove that the SNR is increased by nearly 2.04 dB for single noisy LFM signal and multi-component LFM signal can be separated effectively by using FRFT.Finally,an improved filtering algorithm based on time-delay correlation and FRFT for single noisy LFM signal is proposed.Compared with the traditional FRFT filtering algorithm,the calculation amount of the improved algorithm is reduced.The SNR is increased by nearly 2.41 dB after filtering in the optimal fractional Fourier domain.Theoretical analysis and simulation results show the feasibility and effectiveness of the improved filtering algorithm.