Sparsity-Aware Noise Subspace Fitting for DOA Estimation.

Area of Science:

• Signal Processing
• Array Signal Processing
• Optimization

Background:

• Direction-of-Arrival (DOA) estimation is crucial for sensor array applications.
• Existing methods may require accurate initialization or have high computational costs.
• Sparsity-aware techniques offer potential for improved DOA estimation performance.

Purpose of the Study:

• To develop a novel sparsity-aware algorithm for DOA estimation.
• To formulate the problem as a convex optimization problem for global convergence.
• To enhance solution sparsity and improve estimation resolution.

Main Methods:

• Developed the Sparsity-Aware Noise Subspace Fitting (SANSF) algorithm.
• Formulated the DOA estimation as a convex Linearly Constrained Quadratic Programming (LCQP) problem.
• Incorporated weighted objective function, L1 norm, and non-negative constraints for sparsity.

Main Results:

• The SANSF algorithm achieves global convergence without requiring accurate initialization.
• Demonstrated enhanced resolution in DOA estimation compared to existing sparsity-aware methods.
• Achieved lower computational burden than competing techniques using simulation and real data.

Conclusions:

• The proposed SANSF algorithm offers a computationally efficient and accurate solution for DOA estimation.
• SANSF effectively leverages sparsity for improved resolution in sensor array processing.
• The convex formulation ensures reliable convergence and ease of implementation.