A Fast Space-Time Adaptive Processing Algorithm Based on Sparse Bayesian Learning for Airborne Radar.

Area of Science:

• Radar Systems Engineering
• Signal Processing
• Machine Learning

Background:

• Space-time adaptive processing (STAP) is crucial for airborne radar, but performance degrades in heterogeneous clutter due to insufficient independent and identically distributed (i.i.d.) training samples.
• Existing sparse recovery (SR) methods to mitigate sample limitations often involve high computational complexity.

Purpose of the Study:

• To develop a computationally efficient sparse recovery approach for STAP in airborne radar systems.
• To address the challenge of insufficient training data in heterogeneous clutter environments.

Main Methods:

• A fast group sparse Bayesian learning (SBL) algorithm is proposed, which identifies the data's support space instead of using all dictionary atoms.
• The approach extends hierarchical models to complex-valued signals by treating real and imaginary components as independent real variables.

Main Results:

• The proposed fast group SBL method significantly reduces computational complexity compared to existing SR techniques.
• The algorithm effectively utilizes limited training samples in heterogeneous clutter for improved STAP performance.
• Validation using both simulated and measured data confirms the algorithm's efficiency and effectiveness.

Conclusions:

• The developed fast group sparse Bayesian learning approach offers a computationally efficient solution for STAP in airborne radar.
• This method enhances clutter suppression and moving target detection capabilities, particularly in challenging heterogeneous environments.
• The technique provides a practical advancement for radar signal processing with real-world applicability.