Frequency estimation of chirp signals based on fractional fourier transform combined with Otsu’s method

Abstract

The parameters estimation of linear frequency modulation (LFM) signals, which also called chirp signals, have been widely applied in optical communication, LIDAR and many other fields. Fractional Fourier transform (FRFT) is one of the most important methods for estimating the frequency of chirp signals. However, the traditional estimation methods based on FRFT are incapable of high-precision parameter estimation for chirp signals, especially multi-component LFM signals. Therefore, we present a novel parameter estimation method based on the Otsu ratio, i.e., the interclass ratio of impulse signal to noise. An algorithm is used to determine this ratio, and then the frequency parameters of chirp signals are estimated by using the ratio as the fitness function. The parameter estimation of single-component and multi-component chirp signals is simulated. The results show that the proposed method has higher estimation accuracy than the traditional FRFT method.