Robust Method via Independent Component Analysis with Additive Noise

Yong Jian Zhao; Bo Qiang Liu; Hong Run Wang

doi:10.4028/www.scientific.net/AMR.113-116.272

Paper Titles

Study on Treatment of Simulated Wastewater Containing Phenol with the Peanut Shells Activated Carbon
p.254

Primary Study on the Pollution Control of Overflow by Stepwise Adsorption Purification Bed
p.259

The Game Analysis of Negative Externality of Environmental Logistics and Government Regulation
p.263

Evaluation of Biodegradability of Amine Collectors
p.267

Robust Method via Independent Component Analysis with Additive Noise
p.272

Outside Flow Field Characteristic of Biological Carriers
p.276

The Experiment Study of Nitrobenzene Polluted Groundwater by Catalytic Oxidation
p.280

Electricity Generation from Cattle Dung Using Microbial Fuel Cell in Acidification Process of Two-Phase Anaerobic Digestion System
p.286

The Regulatory Policy Design of Nonpoint Pollution from Agricultural Activities: Issues and Challenges in China
p.291

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 113-116Robust Method via Independent Component Analysis...

Robust Method via Independent Component Analysis with Additive Noise

Abstract:

Blind source separation via independent component analysis (ICA) has received increasing attention because of its potential application in signal processing system. The existing ICA methods can not give a consistent estimator of the mixing matrix because of additive noise. Based on interpretation and properties of the vectorial spaces of sources and mixtures, a new ICA method is presented in this paper that may constructively reject noise so as to estimate the mixing matrix consistently. This procedure may capture the underlying source dynamics effectively even if additive noise exists. The simulation results show that this method has high stability and reliability in the process of revealing the undering group structure of extracted ICA components.