Partitioned factor analysis for interference suppression and source extraction

Authors:
Srikantan S. Nagarajan;Hagai T. Attias;Kensuke Sekihara;Kenneth E. Hild
Affiliations:
Dept. of Radiology, University of California at San Francisco, CA;Golden Metallic Inc., San Francisco, CA;Dept. of Electronic Systems and Engineering, Tokyo Metropolitan Institute of Technology, Japan;Dept. of Radiology, University of California at San Francisco, CA
Venue:
ICA'06 Proceedings of the 6th international conference on Independent Component Analysis and Blind Signal Separation
Year:
2006

Citing 2
Cited 1

Elements of information theory

Elements of information theory
Fast and robust fixed-point algorithms for independent component analysis

IEEE Transactions on Neural Networks

Fetal cardiac signal extraction from magnetocardiographic data using a probabilistic algorithm

Signal Processing

Quantified Score

Hi-index	0.00

Visualization

Abstract

It is common for data to be contaminated with artifacts, interference, and noise. Several methods including independent components analysis (ICA) and principal components analysis (PCA) have been used to suppress these undesired signals and/or to extract the underlying (desired) source waveforms. For some data it is known, or can be extracted post hoc, how to partition the data into periods of source activity and source inactivity. Two examples include cardiac data and data collected using the stimulus-evoked paradigm. However, neither ICA nor PCA are able to take full advantage of the knowledge of the partition. Here we introduce an interference suppression method, partitioned factor analysis (PFA), that takes into account the data partition.