Abstract: For the short-range defensive weapons, extracting the acoustic characteristic of attacking aircraft at low altitude is an effective way to detect and recognize the target. Firstly, the signal characteristics of radiation noise of the typical vehicle propulsion system are analyzed, thus the mathematical model of propulsion system noise and the receiving signal model of passive acoustic early-warning system are given. Furthermore, the Radon transformation based LOFAR spectrogram extraction algorithm is proposed to obtain the rotation speed characteristics of the aircrafts propulsion system. Finally, the feasibility of the proposed algorithm is validated by analyzing the result of the measured aircraft noise.