拖线阵畸变阵形和目标方位估计方法

Distorted Array Shape and Target Azimuth Estimation Algorithm during Towing Platform Maneuvers

摘要:

拖线阵声纳广泛应用于水下目标检测，在舰艇平台转向机动时，柔性的拖线阵不再保持直线形，阵形发生畸变，阵元位置未知，导向矢量失配，进而导致拖线阵声纳性能急剧下降。针对这一问题，本文提出了一种基于稀疏贝叶斯学习(Sparse Bayesian Learning, SBL)的阵形和目标方位估计算法。该算法仅利用接收到的声学信号估计目标方位，将转弯时的畸变阵形建模为抛物线形，并将抛物线参数作为SBL的超参数，实现了目标方位和畸变阵形的同时估计。仿真分析结果表明：在畸变阵形为抛物线模型的假设下，算法可以同时对畸变阵形和目标方位进行估计，并具有更好的目标检测能力和更高的分辨力。

关键词:拖线阵阵形畸变;稀疏贝叶斯学习;畸变阵形估计;目标方位估计;

Abstract:

Towed array sonar is widely used in underwater target detection. During ship platform maneuvering, the towed array shape is distorted, deviating from linear configuration. This results in unknown array elements positions and steering vector mismatch, leading to significant performance degradation of the sonar. To solve this problem, the Shape and Azimuth estimation algorithm based on Sparse Bayesian Learning was proposed. The algorithm estimates target azimuth solely using received acoustic data, and the distorted array shape during maneuvering is modeled as a parabolic curve, with the parabolic curve, with the parabolic parameters treated as hyper-parameters in the SBL framework, enabling simultaneous estimation of both array shape parameters and target DOAs. Simulation results demonstrate that under the assumption of parabolic distortion model, the proposed algorithm achieves concurrent estimation of array distortion and target azimuths, exhibiting enhanced target detection capability and superior resolution.

Key words:towed array distortion;Sparse Bayesian Learning;distorted array shape estimation;target azimuth estimation;