Robust underwater direction-of-arrival tracking based on variational Bayesian extended Kalman filter

Tracking the direction of arrival (DOA) of underwater signals is a key challenge in marine environments. Environmental noise introduces uncertainty that can degrade tracking accuracy. Prior research has shown that conventional DOA methods often fail under such conditions. This gap motivated the development of more robust techniques. No prior work had resolved the simultaneous estimation of both noise and bearing. Existing methods rely on assumptions about noise that may not hold in real-world settings. That uncertainty drives the need for adaptive filters. This paper addresses the problem by integrating Bayesian inference into DOA tracking.

The aim of this study is to develop a reliable DOA tracking method for underwater targets. The focus is on environments where noise is unpredictable and variable. The researchers propose a technique that estimates both the target angle and environmental noise. This approach is intended to improve robustness in real-world conditions. The motivation stems from limitations in traditional DOA methods. The study tests the method using real sea trial data. The goal is to demonstrate superior performance in terms of accuracy and reliability. The proposed solution is designed to adapt to uncertain underwater environments.

The method uses a variational Bayesian extended Kalman filter (VB-EKF) to estimate DOA. This approach combines Bayesian inference with extended Kalman filtering. The VB-EKF allows for simultaneous estimation of noise and bearing angle. The technique is applied to real-world data from the South China Sea. The researchers use sea trial data collected in July 2021 for validation. The method accounts for unknown environmental noise during tracking. The VB-EKF framework enables adaptive noise modeling in real time. The study compares the proposed method to traditional DOA estimation techniques.

The VB-EKF method achieved higher accuracy than traditional DOA techniques. The results show improved robustness in uncertain underwater environments. The study reports that the proposed method outperforms existing methods. The sea trial data confirmed the method's reliability in real-world conditions. The VB-EKF provided consistent tracking results across all test scenarios. The simultaneous estimation of noise and bearing reduced estimation errors. The method demonstrated superior performance in both accuracy and stability. The researchers observed a significant reduction in tracking deviations compared to conventional approaches.

The authors state that the VB-EKF method offers a reliable solution for underwater DOA tracking. They propose that this technique is better suited for environments with uncertain noise. The study concludes that the VB-EKF provides more accurate and robust results. The researchers suggest that the method is effective in real-world marine conditions. The VB-EKF approach is presented as a viable alternative to traditional DOA techniques. The study supports the claim that simultaneous noise and bearing estimation improves tracking. The authors note that the method's performance was confirmed using real sea trial data. They propose that this technique may be useful in future underwater monitoring applications.