Joint Model-Order and Robust DoA Estimation for Underwater Sensor Arrays

Area of Science:

• Signal Processing
• Underwater Acoustics
• Array Signal Processing

Background:

• Direction-of-Arrival (DoA) estimation is crucial for sensor array systems.
• Compressive Sensing (CS) offers superior DoA estimation with limited snapshots compared to conventional methods.
• Underwater acoustic arrays face challenges like unknown source numbers, faulty sensors, and low SNR.

Purpose of the Study:

• To investigate CS-based robust DoA estimation for joint impacts of faulty sensors and low SNR in underwater acoustic arrays.
• To develop a CS technique that does not require prior knowledge of the source number.
• To enhance DoA estimation performance under realistic adverse conditions.

Main Methods:

• Utilized compressive sensing (CS) and sparse reconstruction techniques.
• Developed a modified stopping criterion for the reconstruction algorithm.
• Incorporated faulty sensor information and received SNR into the algorithm.
• Employed Monte Carlo simulations for performance evaluation.

Main Results:

• The proposed CS-based method demonstrates robust DoA estimation under joint faulty sensor and low SNR conditions.
• The technique successfully estimates DoA without prior knowledge of the source number.
• Performance evaluations show advantages over existing methods in challenging underwater scenarios.

Conclusions:

• The developed CS-based approach provides a robust solution for DoA estimation in underwater acoustic systems with faulty sensors and low SNR.
• This method enhances the reliability of target bearing estimation in practical, noisy environments.
• The ability to forgo prior source number knowledge simplifies the application of DoA estimation.