Joint Smoothed l₀-Norm DOA Estimation Algorithm for Multiple Measurement Vectors in MIMO Radar

Area of Science:

• Signal Processing
• Radar Systems
• Array Signal Processing

Background:

• Direction-of-arrival (DOA) estimation is crucial for radar systems, often facing multiple measurement vector (MMV) challenges.
• Existing methods struggle with efficiency and performance in noisy environments.

Purpose of the Study:

• To propose a novel, fast sparse DOA estimation algorithm for MMV in Multiple-Input Multiple-Output (MIMO) radar.
• To improve DOA estimation accuracy and computational efficiency, particularly in the presence of Gaussian noise.

Main Methods:

• A low-complexity, high-order cumulants-based data matrix is generated to mitigate white or colored Gaussian noise.
• A joint smoothed function is designed for the MMV case, forming a joint smoothed l0-norm sparse representation framework.
• Gradient-based sparse signal reconstruction is applied to the MMV-based joint smoothed function for DOA estimation.

Main Results:

• The proposed algorithm achieves fast sparse representation for MMV problems.
• It demonstrates robust performance against both white and colored Gaussian noises.
• The joint algorithm is significantly faster (approximately two orders of magnitude) than l1-norm minimization methods like l1-SVD and RV l1-SRACV.

Conclusions:

• The novel joint smoothed l0-norm algorithm provides a computationally efficient and accurate solution for DOA estimation in MMV scenarios within MIMO radar.
• It offers superior performance compared to existing l1-norm minimization techniques, especially in terms of speed and noise resilience.