Sparse Representations for Structured Noise Filtering

TL;DR

This paper explores the use of sparse representations and an extended Oblique Matching Pursuit technique for structured noise filtering, especially in ill-posed problems, by leveraging orthogonal projections for improved signal discrimination.

Contribution

It introduces an extended approach to Oblique Matching Pursuit that utilizes orthogonal projections to enhance structured noise filtering in ill-posed problems.

Findings

Effective noise filtering in ill-posed problems

Extension of Oblique Matching Pursuit technique

Improved signal discrimination using orthogonal projections

Abstract

The role of sparse representations in the context of structured noise filtering is discussed. A strategy, especially conceived so as to address problems of an ill posed nature, is presented. The proposed approach revises and extends the Oblique Matching Pursuit technique. It is shown that, by working with an orthogonal projection of the signal to be filtered, it is possible to apply orthogonal matching pursuit like strategies in order to accomplish the required signal discrimination