Abstract: Aiming at the problem of Direction-of-Arrival (DOA) estimation of hyperbolic Frequency Modulation (FM) signal in active detection or pulse detection, a multi-signal classification direction finding algorithm based on Parameterized Time-Frequency Transform (PTFT-MUSIC) is proposed. The parameterized transform core of the hyperbolic FM signal is determined by the transmitted signal, and the frequency domain parameterized time-frequency transform of the received signal is carried out. The time-frequency distribution model of the array signal is established by using the obtained time-frequency distribution, and the MUSIC algorithm based on the time-frequency distribution matrix is designed to realize the DOA estimation of the hyperbolic FM signal. Through simulation and experiment, the estimation error and multi-target resolution performance of the algorithm are analyzed. The simulation and sea experiment results show that compared with the existing time-frequency MUSIC algorithm, the PTFT-MUSIC algorithm can effectively improve the spatial-spectral resolution and DOA estimation performance. At the same time, the algorithm has the ability to screen specific FM signals, combined with the time-frequency filtering algorithm, the coherent direct wave interference can be effectively suppressed. When applied to multistatic sonar system, the positioning performance of sonar is improved effectively.