Feng Chen

To reduce the computational burden of conventional spectral search direction-of-arrival (DOA) estimation algorithms for acoustic vector sensor arrays (AVSAs), this paper proposes a low-complexity DOA estimation method based on semi-real-valued noise power invariance (SR-NPI). The proposed method is developed for centrosymmetric AVSAs under the required steering-vector parity and pressure-channel conjugate-symmetry conditions after pressure–velocity (PV) co-processing. The AVSA measurements are first preprocessed through PV co-processing, and a pseudo-data covariance matrix is then reconstructed by exploiting the complex conjugate relationship between the true DOAs and their symmetric virtual DOAs. By introducing a scanning source into the reconstructed covariance matrix, a DOA-dependent spatial spectrum is constructed according to the eigenvalue-ordering behavior. Since the reconstructed matrix contains both the true DOAs and the symmetric virtual DOAs, the angular search range can be reduced to one half of the original domain. Under the tested configuration, the proposed method reduces the computational cost to approximately 1.92% of that of the original NPI algorithm. Simulation and sea trial results indicate that SR-NPI can maintain competitive estimation accuracy while significantly reducing computational complexity under the centrosymmetric-array conditions.

Underwater acoustic target recognition (UATR) is challenged by complex marine noise, scarce labeled data, and inadequate multi-scale feature extraction in conventional methods. This study proposes DART-MT, a semi-supervised framework that integrates a Dual Attention Parallel Residual Network Transformer with a mean teacher paradigm, enhanced by domain-specific prior knowledge. The architecture employs a Convolutional Block Attention Module (CBAM) for localized feature refinement, a lightweight New Transformer Encoder for global context modeling, and a novel TriFusion Block to synergize spectral–temporal–spatial features through parallel multi-branch fusion, addressing the limitations of single-modality extraction. Leveraging the mean teacher framework, DART-MT optimizes consistency regularization to exploit unlabeled data, effectively mitigating class imbalance and annotation scarcity. Evaluations on the DeepShip and ShipsEar datasets demonstrate state-of-the-art accuracy: with 10% labeled data, DART-MT achieves 96.20% (DeepShip) and 94.86% (ShipsEar), surpassing baseline models by 7.2–9.8% in low-data regimes, while reaching 98.80% (DeepShip) and 98.85% (ShipsEar) with 90% labeled data. Under varying noise conditions (−20 dB to 20 dB), the model maintained a robust performance (F1-score: 92.4–97.1%) with 40% lower variance than its competitors, and ablation studies validated each module’s contribution (TriFusion Block alone improved accuracy by 6.9%). This research advances UATR by (1) resolving multi-scale feature fusion bottlenecks, (2) demonstrating the efficacy of semi-supervised learning in marine acoustics, and (3) providing an open-source implementation for reproducibility. In future work, we will extend cross-domain adaptation to diverse oceanic environments.