Radar One-Dimensional Range Profile Imaging Using SBL Algorithm and Kalman Filter Compressed Sensing

In this paper, the authors implement radar one-dimensional range profile imaging using both the Sparse Bayesian Learning (SBL) algorithm and a Kalman filter-based compressed sensing approach. The SBL algorithm operates by iteratively estimating hyperparameters to recover sparse signals through Bayesian inference, typically implemented with expectation-maximization iterations. The Kalman filter compressed sensing method employs state-space modeling where the Kalman filter recursively estimates sparse coefficients while handling measurement noise. These techniques provide richer information and higher precision for radar image processing and analysis. Following imaging performance analysis of both algorithms, the authors derived significant conclusions about their relative computational efficiency, reconstruction accuracy under different signal-to-noise ratios, and sparse recovery capabilities. Through in-depth study of these findings, researchers can further refine and optimize the practical application of these techniques to better meet real-world radar imaging requirements.