Skip to content

thanhle88/aOPIT

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

12 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

aOPIT: A Robust Sparse Subspace Tracking Method Using Alpha Divergence With Non-Gaussian Noises

Subspace tracking is a fundamental problem in signal processing, where the goal is to estimate and track the underlying subspace that spans a sequence of data streams over time. In high-dimensional settings, data samples are often corrupted by non-Gaussian noises and may exhibit sparsity. This paper explores the alpha divergence for sparse subspace estimation and tracking, offering robustness to data corruption.

The proposed algorithm, termed aOPIT, is a robust variant of our OPIT method (Thanh et al., IEEE TSP 2024) using alpha divergence. aOPIT outperforms the state-of-the-art robust subspace tracking methods for the task of dealing with mixtures of noises while achieving a low computational complexity and memory storage.

Demo

Please run

  • test_alpha_OPIT.m: To illustrate the performance of aOPIT in comparsion with two SOTA algorithms, including aFAPI and OPIT.

Reference

This code is free and open source for research purposes. If you use this code, please acknowledge the following paper.

[1] T.G.T. Loan#, N.H. Lan#, N.T.N. Lan#, D.H. Son, T.T.T. Quynh, K. Abed-Meraim, N.L. Trung, L.T. Thanh. "Robust Sparse Subspace Tracking from Corrupted Data Observations". Proc. IEEE ISCIT, 2025.

About

[ISCIT 2025] aOPIT: Robust Sparse Subspace Tracking from Corrupted Data Observations. In 2025 24th IEEE International Symposium on Communications and Information Technologies

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors

Languages