A fast algorithm for parameter estimation of multi-component LFM signal at low SNR
Pu Wang, Jianyu Yang, Yuming Du · 2005
A fast method for parameter estimation of the multi-component linear frequency modulated (multi-LFM) signal is proposed. The signal detector and chirp rate estimator are based on the modulus square of the fractional autocorrelation. By using the estimated chirp rate, an approach based on the fractional Fourier transform (FrFT) is employed for estimation of the amplitude and centre frequency. The proposed algorithm reduces two-dimensional searches, widely used in the time-frequency based method, FrFT and chirp Fourier transform, into two one-dimensional searches. By utilizing the discrete FrFT, along with the fast Fourier transform (FFT) algorithm, the proposed method is a computationally fast alternative for LFM signal detection and parameter estimation. Analysis of the multi-LFM signal is performed using the CLEAN technique as well. Finally, computer simulations are provided to illustrate the performance of the proposed algorithm.