Title: Adaptive fast algorithm based on natural gradient for instantaneous blind source separation

Authors: Niva Das; Binodini Tripathy; Srikanta Patnaik

Addresses: Department of Electronics and Communication Engineering, SOA University, Jagmohan Nagar, Bhubaneswar-751030, Odisha, India ' ETC Department, KIIT University, Campus-2, Kolab Campus, Bhubaneswar-751024, Odisha, India ' Department of Computer Science and Engineering, SOA University, Jagmohan Nagar, Bhubaneswar-751030, Odisha, India

Abstract: A fast adaptive algorithm based on natural gradient is proposed in this paper to address the convergence issue in instantaneous blind source separation (BSS) problem in a noisy and non-stationary mixing scenario. The natural gradient technique overcomes slow convergence properties of the gradient adaptation technique when the slope of the cost function varies widely for small changes in the parameters. To speed up the convergence further we adopt one of the heuristic methods by incorporating a momentum term. Synthetically generated data as well as real world data have been considered for validation purpose. Numerical experiments on sinusoidal signals and acoustic electromechanical signals confirm the superior performance of the proposed algorithm over the conventional natural gradient algorithm (NGA) in both noisy and noiseless situation as well as in stationary and non-stationary mixing scenario.

Keywords: blind source separation; instantaneous BSS; natural gradient; non-stationary systems: independent component analysis; ICA; sinusoidal signals; acoustic electromechanical signals.

DOI: 10.1504/IJICT.2016.076759

International Journal of Information and Communication Technology, 2016 Vol.8 No.4, pp.307 - 314

Received: 26 Oct 2013
Accepted: 15 Jan 2014

Published online: 26 Mar 2016 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article