Study of Sparsity-Aware Set-Membership Adaptive Algorithms with Adjustable Penalties

TL;DR

This paper introduces adaptive filtering algorithms that automatically adjust penalty parameters to exploit sparsity, leading to faster convergence in system identification tasks.

Contribution

We develop a framework for automatically tuning penalty functions and step sizes in sparsity-aware adaptive algorithms, improving their performance over fixed-parameter methods.

Findings

Proposed algorithms outperform existing methods in convergence speed.

Automatic adjustment of penalties enhances sparsity exploitation.

Framework applicable to various adaptive filtering scenarios.

Abstract

In this paper, we propose sparsity-aware data-selective adaptive filtering algorithms with adjustable penalties. Prior work incorporates a penalty function into the cost function used in the optimization that originates the algorithms to improve their performance by exploiting sparsity. However, the strength of the penalty function is controlled by a scalar that is often a fixed parameter. In contrast to prior work, we develop a framework to derive algorithms that automatically adjust the penalty function parameter and the step size to achieve a better performance. Simulations for a system identification application show that the proposed algorithms outperform in convergence speed existing sparsity-aware algorithms.