An Optimal Subspace Deconvolution Algorithm for Robust and High-Resolution Beamforming

Area of Science:

• Signal Processing
• Array Signal Processing
• Acoustic Signal Processing

Background:

• Direction of Arrival (DOA) estimation is crucial in various applications.
• High-resolution DOA estimation methods often struggle with robustness in noisy environments.
• Conventional methods require high signal-to-noise ratios (SNR), limiting their applicability.

Purpose of the Study:

• To propose a novel Subspaces Deconvolution Vector (SDV) beamforming method.
• To enhance the robustness of high-resolution DOA estimation in noisy conditions.
• To improve the separation of signals from noise for accurate DOA estimation.

Main Methods:

• Transferred DOA estimation to a spatial sample classification problem.
• Utilized the difference in phase and power spectrum between signals and noise.
• Developed the Subspaces Deconvolution Vector (SDV) beamforming approach.
• Employed incoherent eigenvalues in the frequency domain for initial beamforming.
• Combined subspace deconvolution vectors with conventional beamforming principles.

Main Results:

• The SDV method achieves high-resolution DOA estimation with improved robustness.
• Effective signal-noise separation was demonstrated even at input SNRs below -27 dB.
• Clear angle estimations were obtained by the SDV method under low SNR conditions.
• Successful application in a marine background for noise separation and signal characteristic recruitment.

Conclusions:

• The SDV beamforming method offers a robust solution for high-resolution DOA estimation.
• The method excels in challenging environments with significant noise.
• SDV enhances accuracy and stability in DOA estimation, particularly at low SNRs.