A distributed subband valley fusion (DSVF) method for low frequency broadband target localization

Area of Science:

• Signal Processing
• Acoustic Localization
• Multi-Array Systems

Background:

• Traditional bearings-only target localization (BOTL) methods often involve complex two-step processing.
• Accurate target localization in multi-array systems remains a challenge, especially in complex environments.

Purpose of the Study:

• To propose a distributed subband valley fusion method for direct target localization.
• To enhance localization accuracy and address data association in multi-target scenarios.

Main Methods:

• Directly obtaining target location estimates from the intersection of bearing lines from each array.
• Utilizing subband valley energy detection on the reciprocal of steered response power.
• Employing bearing information across all frequencies and ranges to generate bearing lines and intersection structures.
• Solving data association in multi-target situations by distinguishing true targets from ghost targets based on power and bearing line counts.

Main Results:

• Achieved higher localization accuracy compared to traditional BOTL methods.
• Demonstrated effective handling of multiple targets and ghost targets.
• Validated the method's performance using the SWellEx-96 shallow water experiment data set.

Conclusions:

• The proposed distributed subband valley fusion method offers a more accurate and direct approach to target localization.
• The method shows significant potential for applications in sonar and other multi-array sensing systems.