Comparative Study of Compressed Sensing Methods for Spherical Near-Field Antenna Measurements with Non-Uniform Sparse Sampling

This paper presents a comprehensive comparative study of three compressed sensing algorithms-Basis Pursuit (BP), Orthogonal Matching Pursuit (OMP), and Two-Step Iterative Shrinkage/Thresholding (TwIST)-for near-field to far-field transformation (NFFFT) under different sampling schemes. The evaluation focuses on their performance in reconstructing antenna radiation patterns in the XOZ and YOZ planes, with particular attention to maximum and mean pattern errors. Numerical results demonstrate that TwIST achieves superior accuracy with minimal artifacts, outperforming BP by a slight margin, while OMP exhibits the highest reconstruction errors due to its sensitivity to sampling conditions. The study highlights the critical trade-offs between computational efficiency and reconstruction fidelity in antenna measurements. These findings provide practical insights for selecting appropriate algorithms in NFFFT applications, especially when optimizing for accuracy under limited sampling conditions.