A Method of Radar Signal Detection Based on Nonlinear Square Short-Time Fourier Transform

The frequency measurement of radar signals is one of the core functions of electronic reconnaissance receivers. In this paper, a nonlinear squared short-time Fourier transform frequency measurement algorithm (NSSTFT) is proposed for the problem of low accuracy and easy spectrum leakage of single-pulse radar signals measured by the traditional short-time Fourier transform algorithm (STFT). The algorithm firstly obtains the frequency difference information and the number of targets of the input signal by performing nonlinear square transformation of the input signal, and the length of the FFT can be determined adaptively from the frequency difference information. Then this paper proposes a frequency-weighted fusion algorithm for the frequency points where the spectrum leakage occurs, which is used to estimate the actual signal frequency. Simulation experiments show that the NSSTFT algorithm proposed in this paper can reduce the distortion effect caused by spectrum leakage and effectively improve the frequency measurement accuracy of electronic reconnaissance receivers.