Coherent multi-beam ray-based blind deconvolution with curvature compensation for channel impulse response...

Area of Science:

• Signal Processing
• Array Signal Processing
• Electromagnetics

Background:

• Ray-based blind deconvolution methods for channel impulse response (CIR) estimation suffer from degradation due to near-field propagation effects.
• Array deformation in sensor arrays introduces significant phase errors, complicating accurate CIR estimation.
• Existing methods struggle to effectively mitigate these issues simultaneously.

Purpose of the Study:

• To develop a robust method for mitigating degradation in ray-based blind deconvolution for CIR estimation.
• To address the challenges posed by near-field propagation and array deformation.
• To improve the accuracy and reliability of channel state information in complex environments.

Main Methods:

• Utilizes sparse Bayesian learning to identify and locate dispersed beam peaks.
• Employs coherent combination of per-beam CIRs to sharpen the target wavefront.
• Estimates inter-element delays by extracting dominant arrival times from combined CIRs.
• Constructs a curvature-compensated steering vector to correct phase errors and recover multipath structure.

Main Results:

• Successfully mitigates degradation in ray-based blind deconvolution for CIR estimation.
• Effectively handles near-field propagation and array deformation.
• Recovers multipath structure by removing phase errors introduced by array deformation.
• Demonstrates improved performance through Monte Carlo simulations and validation with at-sea data.

Conclusions:

• The proposed method offers a significant improvement for CIR estimation in challenging near-field and array deformation scenarios.
• Accurate multipath recovery and phase error correction are achieved.
• The approach is validated and shows promise for practical applications in wireless communications and sensing.