Year: 2025
Type: article
Abstract: Addressing the issue of poor positioning performance of the near-field MVDR (Minimum Variance Distance Reciprocal) algorithm due to limited sampling data, this study proposes a combination of vector hydrophones and an improved angle loading technique to enhance the robustness of the algorithm. The reasons for the degradation of positioning performance of the MVDR algorithm with a small number of snapshots are analyzed from two perspectives: optimal weight vector and output power spectrum. Firstly, the traditional angle loading algorithm is utilized to reduce the degree of noise feature diffusion, and then the loaded feature values are squared to obtain an improved covariance matrix, effectively resolving the issue of resolution degradation associated with traditional angle loading methods. Finally, the effectiveness and stability of the algorithm are verified through simulations and experimental data. The results demonstrate that the improved angle loading algorithm proposed in this paper performs comparably to traditional methods but significantly enhances resolution, making it more suitable for the positioning of underwater acoustic near-field targets.
Source: Scientific Reports
Authors: Qing Ji, Dawei Xiao, Lulu Du, Taotao Xie, Yandong Pang
Language: English
Subfield: Signal Processing
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $2,690
Award: G2294522191
OpenAlex ID: W4415620624