A Sensing and Tracking Algorithm for Multiple Frequency Line Components in Underwater Acoustic Signals.

Area of Science:

• Signal Processing
• Acoustics
• Machine Learning

Background:

• Underwater acoustic signal analysis often struggles with detecting and tracking multiple weak or time-varying frequency components.
• Existing methods may face challenges in low signal-to-noise ratio (SNR) environments and with complex signal patterns.

Purpose of the Study:

• To develop an automated method for reliable and efficient sensing and tracking of multiple frequency lines in underwater acoustic signals.
• To improve the detection of weak and time-varying frequency components in lofargrams, particularly under low SNR conditions.

Main Methods:

• A novel approach using Hidden Markov Models (HMM) for automatic detection and tracking of multiple frequency lines in lofargrams.
• Segmentation of lofargrams into sub-lofargrams for targeted screening and HMM-based detection.
• Image stitching and statistical modeling for merging frequency lines detected across different sub-lofargrams.

Main Results:

• The proposed HMM-based method effectively detects multiple time-varying frequency lines in underwater acoustic signals.
• The algorithm demonstrates robust performance even under low signal-to-noise ratio (SNR) conditions.
• Significant reduction in computational load compared to traditional methods.

Conclusions:

• The developed algorithm offers enhanced multiple frequency line sensing abilities for underwater acoustic signals.
• The method provides a computationally efficient technique for feature sensing and tracking.
• Potential applications include unattended equipment like sonar and submerged buoys.