On the Kalman Filter Parameter Estimation Methods for Blind Source Separation

Authors

DOI:

https://doi.org/10.14209/jcis.2025.1

Keywords:

blind source separation, Kalman filtering, parameter estimation, joint estimation, dual estimation, signal processing

Abstract

Blind Source Separation (BSS) is a well-known problem in signal processing and still receives attention from the scientific community, given its applicability in different areas. This work presents a theoretical background overview of the Kalman Filter formulation and its applicability to the BSS problem as a parameter estimator in two different approaches: joint (JEKF) and dual parameter estimation (DEKF). These approaches are evaluated in different scenarios for first-order autoregressive source signals, with analysis of the initialization details, presenting simulation results and performance comparison with classic algorithms, SOBI and SONS, evaluated by SIR, MER and MSE. The results showed that both the JEKF and DEKF algorithms can perform separation in a two-source-two-mixture scenario. In general, over the scenarios studied, DEKF presented a better performance when compared to the JEKF on the evaluated metrics. However, neither algorithm correctly estimated the parameters for mixtures involving more than two sources, showing convergence issues and sensitivity to initialization for an increased number of sources.

Downloads

Download data is not yet available.

Author Biographies

Alexandre Miccheleti Lucena, Universidade Federal do (UFABC)

Alexandre Miccheleti Lucena is currently pursuing a Ph.D. in the Information Engineering postgraduate program at the Federal University of ABC (UFABC). He earned his Master's degree in Information Engineering in 2021, following his graduation in Information Engineering from the Federal University of ABC in 2019. He holds a bachelor's degree in Science and Technology from the same university in 2017, including a sandwich period at Shibaura Institute of Technology in Japan in 2015. His expertise lies in digital signal processing, with a focus on multimedia signals, blind source separation and machine learning.

Kenji Nose-Filho, Universidade Federal do (UFABC)

Kenji Nose Filho is currently an Assistant Professor at the Federal University of ABC (UFABC). He received his B.E. and M.E. degrees in Electrical Engineering from UNESP, Ilha Solteira campus, in 2008 and 2011. He pursued a 6 month exchange at Universidad ORT, Uruguay. Obtaining his Ph.D. at University of Campinas (UNICAMP), he conducted a sandwich Ph.D. at GIPSA-Lab, Grenoble, France. From September 2015 to January 2017 he was a Post-Doc at DSPCom, at UNICAMP. His expertsie lies in adaptive filtering, predictive deconvolution, blind source separation, machine learning, and optimization methods, applied to reflection seismic, time-series analysis, forecasting, and audio and image processing.

Ricardo Suyama, Universidade Federal do (UFABC)

Ricardo Suyama is currently an Associate Professor at the Federal University of ABC (UFABC). He received his B.S., M.S., and Ph.D. degrees in Electrical Engineering from the University of Campinas in 2001, 2003, and 2007, respectively. His expertise lies in Digital Signal Processing, with a focus on the application of Artificial Intelligence in signal processing, Adaptive Filtering, Signal Enhancement, and Source Separation. He is a researcher of the Laboratory of Signals and Systems (LSS) at UFABC.

Downloads

Published

2025-03-24

How to Cite

Miccheleti Lucena, A., Nose-Filho, K., & Suyama, R. (2025). On the Kalman Filter Parameter Estimation Methods for Blind Source Separation. Journal of Communication and Information Systems, 40(1). https://doi.org/10.14209/jcis.2025.1

Issue

Section

Regular Papers
Received 2024-07-26
Accepted 2025-03-01
Published 2025-03-24

Most read articles by the same author(s)