High-resolution sonar imaging based on wideband fast iterative shrinkage thresholding deconvolution beamforming.

Area of Science:

• Underwater Acoustic Imaging
• Signal Processing
• Array Signal Processing

Background:

• Conventional beamforming (CBF) is robust and simple for underwater acoustic imaging but limited by the Rayleigh resolution threshold.
• Existing deconvolution beamforming methods improve resolution but struggle with incoherent targets and narrowband signals, not matching practical sonar models.
• Intensity-based methods fail with coherent targets due to cross-term interference and are unsuitable for wideband signals.

Purpose of the Study:

• To develop a high-resolution imaging method for wideband coherent signals in underwater acoustics.
• To extend deconvolution beamforming capabilities beyond the limitations of intensity-based methods.
• To address the mismatch between current deconvolution techniques and practical sonar imaging models.

Main Methods:

• Introduced a matrix transformation technique to focus wideband signals to a reference frequency, approximating them as narrowband.
• Developed a complex-domain deconvolution beamforming algorithm (MFISTA) to mitigate cross-term interference from coherent targets.
• Utilized simulations with uniform linear arrays and lake-based experiments for validation.

Main Results:

• The proposed MFISTA method achieved a main lobewidth less than 1/4 of CBF and 1/3 of other deconvolution methods.
• Sidelobe levels were reduced by over 5 dB compared to other methods.
• Lake-based experiments confirmed superior resolution, main lobewidth, sidelobe level, and noise immunity over intensity-based methods.

Conclusions:

• The matrix transformation and MFISTA approach effectively extends deconvolution beamforming to wideband coherent signals for high-resolution underwater acoustic imaging.
• This method overcomes limitations of intensity-based techniques, offering significant improvements in resolution and performance.
• The validated performance demonstrates its potential for practical sonar imaging applications.