Remote passive sonar detection with underwater acoustic vector sensor (UAVS) has attracted increasing attention due to its merit in measuring the full sound field information. However, the accurate estimation of the direction-of-arrival (DOA) remains a challenging problem, especially under low signal-to-noise ratio (SNR) conditions. In this paper, a novel convolution (COV)-based single-vector acoustic preprocessing method is proposed on the basis of the single-vector acoustic preprocessing model. In view of the theoretical analysis of the classical single-vector acoustic DOA estimation method, the principle of preprocessing can be described as “to achieve an improved denoising performance in the constraint of equivalent amplitude gain and phase response.” This can be naturally guaranteed by our proposed COV method. In addition, the upper bound with matched filtering (MF) preprocessing is provided in the consideration of the optimal linear signal processing for weak signal detection under Gaussian noise. Numerical analyses demonstrate the effectiveness of our proposed preprocessing method with both vector array signal processing-based and intensity-based methods. Experimental verification conducted in South China Sea further verifies the effectiveness of our approach for practical applications. This work can lay a solid foundation in improving underwater remote vector acoustic DOA estimation under low SNR, and can provide important guidance for future research work.

Very-low-frequency (VLF) passive sonar detection is one of the core technologies for maritime surveillance, although its performance is often severely affected by strong impulsive ocean ambient noise interference. This paper, for the first time, proposes a convolutional neural network (CNN) detection framework with hybrid Log-Mel spectrogram (Log-Mel) and Harmonic–Percussive Source Separation (HPSS) preprocessing. Aiming to highlight the detailed features of low frequencies in accordance with impulsive noise interference removal, the network was trained on a measured dataset in the South China Sea for a whole week by maximize the area under receiver operating characteristic curve (AUC) that corresponds to a false alarm probability of less than 0.1. The test results show that compared with a typical Short-Time Fourier Transform (STFT) input feature, the utilization of Log-Mel and HPSS can be superior, especially utilizing Log-Mel and HPSS(H) features at the same time. Validation with a set of measured moving ship data shows that the detection performance of the proposed hybrid Log-Mel and HPSS-aided CNN can be stable and significantly improve the remote passive sonar detection performance.