Comparison of threshold-based algorithms for sparse signal recovery

TL;DR

This paper compares threshold-based algorithms like Orthogonal Matching Pursuit, Iterative Hard Thresholding, and Single Iteration Reconstruction for sparse signal recovery within compressive sensing, focusing on error and efficiency.

Contribution

It provides a comparative analysis of commonly used threshold-based algorithms for sparse signal reconstruction, highlighting their performance differences.

Findings

OMP has lower reconstruction error than IHT and SIR.

IHT offers faster execution times.

SIR provides a balance between accuracy and speed.

Abstract

Intensively growing approach in signal processing and acquisition, the Compressive Sensing approach, allows sparse signals to be recovered from small number of randomly acquired signal coefficients. This paper analyses some of the commonly used threshold-based algorithms for sparse signal reconstruction. Signals satisfy the conditions required by the Compressive Sensing theory. The Orthogonal Matching Pursuit, Iterative Hard Thresholding and Single Iteration Reconstruction algorithms are observed. Comparison in terms of reconstruction error and execution time is performed within the experimental part of the paper.