Feature Extraction from Passive Sonar Data Based on the Combination of Cepstrum and Short-Time Fourier Transform

Recently, different methods have been proposed for classifying and extracting features from passive sonar data. The main purpose of this paper is to present a feature extraction method from passive sonar data in the presence of noise. Then, the extracted features are used in the classification system and lead to increased efficiency of the classification system compared to previous works. In this paper, the features of the time and frequency domain have been used together. Short-time Fourier transforms with the cepstrum method are used to extract the frequency and time-domain features. The short-time Fourier transform method is used to extract the frequency and time domain features and the cepstrum method is used to extract the features of the time domain. The proposed method was evaluated and tested on the actual collected data of several vessels in the Caspian Sea and the results of the simulations show an improvement in the performance of the proposed method compared to other methods.