Title :
A New Constrained Independent Component Analysis Method
Author :
Huang, De-Shuang ; Mi, Jian-Xun
Author_Institution :
Chinese Acad. of Sci., Hefei
Abstract :
Constrained independent component analysis (cICA) is a general framework to incorporate a priori information from problem into the negentropy contrast function as constrained terms to form an augmented Lagrangian function. In this letter, a new improved algorithm for cICA is presented through the investigation of the inequality constraints, in which different closeness measurements are compared. The utility of our proposed algorithm is demonstrated by the experiments with synthetic data and electroencephalogram (EEG) data.
Keywords :
electroencephalography; independent component analysis; augmented Lagrangian function; constrained independent component analysis method; electroencephalogram data; inequality constraints; negentropy contrast function; synthetic data; Biomedical measurements; Content addressable storage; Data mining; Electroencephalography; Independent component analysis; Information analysis; Laboratories; Lagrangian functions; Machine intelligence; Signal processing algorithms; Constrained optimization; ICA with reference (ICA-R); electroencephalogram (EEG); independent component analysis (ICA); Algorithms; Artificial Intelligence; Computer Simulation; Diagnosis, Computer-Assisted; Electroencephalography; Models, Neurological; Pattern Recognition, Automated; Principal Component Analysis;
Journal_Title :
Neural Networks, IEEE Transactions on
DOI :
10.1109/TNN.2007.895910
