A deep transfer learning for underwater source ranging is proposed, which migrates the predictive ability obtained from synthetic environment (source domain) into an experimental sea area (target domain). A deep neural network is first trained on large synthetic datasets generated from historical environmental data, and then part of the neural network is refined on collected data set for source ranging. Its performance is tested on a deep-sea experiment through comparing with convolutional neural networks of different training datasets. Data processing results demonstrate that the ranging accuracy is considerably improved by the proposed method, which can be easily adapted for related areas.

For a narrowband signal, an oscillating interference pattern is formed with a target's moving when receiving at the bottom of the sea. In this Letter, the interference pattern of a narrowband source is observed using a single vector sensor (SVS). A passive depth estimation method employing a SVS is proposed. This approach processes the signals after the adaptive line enhancing and extracts the vector intensity, which oscillates periodically with the vertical azimuth. The passive estimation is achieved based on the Fourier-transform relationship between the depth and interference period. The simulation and sea experiment verify this method.

To scan the horizontal range of the deep-sea shadow zone with high accuracy and efficiency, enabling active target detection, a horizontal detection range estimation method based on phased sound focusing technology is presented. The model of the phased emission sound field is derived, along with the theoretical estimation formula for the horizontal detection range, by combining the half-power beamwidth of the beam curve at the phased-emitter end with the ray propagation law of deep-sea acoustics. The corresponding relationships among each discrete phased angle, the horizontal distance, and the detection range of the focused beam boundary are detailed. The estimation formula for the horizontal detection range is then improved to account for the sea surface reflection effect, and angle optimization is applied to the transmitted beam curve to counteract the boundary error accumulation effect of the focused sound beam as the sound ray propagates. To address the issue of reduced scanning efficiency at large phased angles, the grating lobe-focused sound beam generated by phased emission array parameters is used to achieve angle scanning within the close deep-sea sound shadow zone, increasing the phased angle scanning efficiency. Finally, the deep-sea phased emission active target detection experiments were carried out, and the echo signal generated by the phased focused sound beam and received by the vertical receiving array was recorded. Numerical simulation and echo signal processing show that the high-efficiency angle scanning of the entire deep-sea sound shadow zone is achieved by the proposed estimation method.

Performance of Doppler sonar is degraded by beam cross coupling errors. This performance degradation presents itself as the loss of precision and bias of velocity estimates output by the system. A model is proposed here to reveal the physical essence of the beam cross coupling. Specifically, the model can analyze the effects of environmental conditions and vehicle attitude on the coupling bias. Based on this model, a phase assignment method is also proposed to reduce the beam cross coupling bias. The results obtained for various settings validate the efficacy of the proposed method.

The relationship between modal elevation angle and the relative arrival time between modes, derived from exploiting modal dispersion, provides source information that is less susceptible to environmental influences. However, the standard method based on modal dispersion has limitations for application. To overcome this, we propose a hybrid method for passive source ranging of low-frequency underwater acoustic-pulse signals in a range-independent shallow-water waveguide. Our method leverages deep learning, utilizing the intermediate results from the standard method as inputs, and short-time conventional beamforming to transform signals received by a vertical line array into a beam-time-domain sound-intensity map. The source range is estimated using an attention-based regression model with a ResNet backbone that has been trained on the beam-time-domain sound-intensity map. Our experimental results demonstrate the superiority of the proposed method, with a mean relative-error reduction of 71%, mean root-squared error reduction of 2.25 km, and an accuracy of 85%, compared to matched-field processing.

The prediction of ocean ambient noise is crucial for protecting the marine ecosystem and ensuring communication and navigation safety, especially under extreme weather conditions such as typhoons and strong winds. Ocean ambient noise is primarily caused by ship activities, wind waves, and other factors, and its complexity makes it a significant challenge to effectively utilize limited data to observe future changes in noise energy. To address this issue, we have designed a multi-modal linear model based on a “decomposition-prediction-modal trend fusion-total fusion” framework. This model simultaneously decomposes wind speed data and ocean ambient noise data into trend and residual components, enabling the wind speed information to effectively extract key trend features of ocean ambient noise. Compared to polynomial fitting methods, single-modal models, and LSTM multi-modal models, the average error of the relative sound pressure level was reduced by 1.3 dB, 0.5 dB, and 0.3 dB, respectively. Our approach demonstrates significant improvements in predicting future trends and detailed fittings of the data.