Accurate broadband modeling of underwater acoustic channels is vital for underwater acoustic detection, localization, and communication. Conventional modeling methodologies, based on methods such as the finite element method, finite difference method, and boundary element method, generally facilitate computation for only a single frequency at a time. However, in broadband modeling, this characteristic presents limitations, requiring multiple computations across frequencies, thereby leading to significant time challenges. To solve this problem, we propose a rapid broadband modeling approach using physics-informed neural networks. By integrating the modal equation of normal modes as a regularization term within the neural network's loss function, the method can achieve rapid broadband modeling of underwater acoustic channel with a sparse set of frequency sampling points. Operating in range-independent underwater environments with a liquid semi-infinite seabed, the method proficiently predicts the channel response across the frequency band from 100 to 300 Hz. Compared to the results obtained from KRAKEN, our method improves computational speed by a factor of 25 at a propagation distance of 20 km, while maintaining a mean absolute error of 0.15 dB for the acoustic channel response.

To acquire the enhanced underwater ship-radiated noise signal in the presence of array shape distortion in a passive sonar system, the phase difference of the line-spectrum component in ship-radiated noise is often exploited to estimate the time-delay difference for the beamforming-based signal enhancement. However, the time-delay difference estimation performance drastically degrades with decreases of the signal-to-noise ratio (SNR) of the line-spectrum component. Meanwhile, although the time-delay difference estimation performance of the high-frequency line-spectrum components is generally superior to that of the low-frequency one, the phase difference measurements of the high-frequency line-spectrum component often easily encounter the issue of modulus 2$\pi$ ambiguity. To address the above issues, a novel time-frequency joint time-delay difference estimation method is proposed in this paper. The proposed method establishes a data-driven hidden Markov model with robustness to phase difference ambiguity by fully exploiting the underlying property of slowly changing the time-delay difference over time. Thus, the phase difference measurements available for time-delay difference estimation are extended from that of low-frequency line-spectrum components in a single frame to that of all detected line-spectrum components in multiple frames. By jointly taking advantage of the phase difference measurements in both time and frequency dimensions, the proposed method can acquire enhanced time-delay difference estimates even in a low SNR case. Both simulation and at-sea experimental results have demonstrated the effectiveness of the proposed method.

Wideband sparse spatial spectrum estimation is an important direction-of-arrival (DOA) estimation method that can obtain a high resolution with few snapshots and a low signal-to-noise ratio. However, in an underwater strong interference environment, the accuracy of DOA estimation may be seriously affected, and even the weak targets could be completely masked. In this paper, we propose a fast matrix filter design method based on truncated nuclear norm regularization to attenuate strong interferences while passing weak targets. The matrix filter operator and the exact covariance matrix after filtering can be obtained simultaneously by solving a convex optimization problem that contains the output power term and non-Toeplitz error propagation control term. Then the modified sparse spectrum fitting algorithm based on the matrix filter is used to estimate spatial spectrum over closely spaced wideband signals. Compared with existing methods, the proposed method achieves higher DOA estimation accuracy and lower computational time for matrix filter design. Meanwhile, the estimation accuracy of the proposed method is verified with the experimental results.

The low-frequency narrowband components (known as lines) in the radiated noise of underwater acoustic targets are an important feature of passive sonar detection. Conventional adaptive line enhancer (ALE) based on the least mean square algorithm has limited performance under colored background noise and low signal-to-noise ratio (SNR). In this paper, by combining the frequency domain sparse model of lines and maximum correntropy criterion (MCC), a β-adaptive l₀-MCC-ALE is proposed to solve the above-mentioned problem. The proposed ALE uses a sparse-driven MCC algorithm to update the weight vector in the frequency domain to further suppress the colored background noise. For the problem that the value of parameter β is sensitive to the performance, β is updated adaptively according to the frequency response of ALE in each iteration. Simulation and real data processing results show that the proposed ALE is insensitive to the given parameter β and has excellent performance for line enhancement. Compared with conventional ALE, the SNR of lines can be improved by 7~8 dB.

Underwater acoustic networks (UANs) in deep-sea suffer from long propagation delays and severe multipath interference under limited bandwidth. This paper presents DNC-CS-TDMA MAC protocol, a topology-transparent hybrid MAC that overlays delay-aware, locally ordered carrier sensing on a TDMA backbone. Each owner slot admits a dynamic neighbor cluster for safe intra-slot reuse via an adaptive CS-Window and reverberation-aware guarding. Without global link-state knowledge, DNC-CS-TDMA selects the active cluster by maximizing an estimated throughput increment that accounts for contention gain and next-slot penalty. Simulations over large-scale, deep-sea topologies show consistent gains in throughput and end-to-end delay over TT-TDMA and DOTS under medium-to-high traffic loads.