Robust common spatial patterns with sparsity. (April 2016)

• Record Type: Journal Article
• Authors: Li, Xiaomeng; Lu, Xuesong; Wang, Haixian
• Is Part Of: Biomedical signal processing and control. Volume 26(2016)
• Page Start: 52
• Page End: 57
• Publication Date: 2016-04
• Subjects: Brain–computer interface (BCI) -- Common spatial patterns -- Robust feature extraction -- Sparsity regularization; Signal processing -- Periodicals; Biomedical engineering -- Periodicals; Signal Processing, Computer-Assisted -- Periodicals; Image Processing, Computer-Assisted -- Periodicals; Biomedical Engineering -- Periodicals; 610.28
• DOI: 10.1016/j.bspc.2015.12.005
• Languages: English
• ISSNs: 1746-8094

Abstract

Highlights: A new approach, termed as sp-CSPL1, is proposed for single-trial EEG classification. L1-norm is used as sparsity regularization, as well as robust modeling. A theoretically validated iterative algorithm is presented. Abstract: Robust and sparse modeling are two important issues in brain–computer interface systems. L1-norm-based common spatial patterns (CSP-L1) method is a recently developed technique that seeks robust spatial filters by using L1-norm-based dispersions. However, the spatial filters obtained are still dense, and thus lack interpretability. This paper presents a regularized version of CSP-L1 with sparsity, termed as sp-CSPL1. It produces sparse spatial filters, which eliminate redundant channels and retain meaningful EEG signals. The sparsity is induced by penalizing the objective function of CSP-L1 with the L1-norm. The sp-CSPL1 approach uses the L1-norm twice for inducing sparsity and defining dispersions simultaneously. The presented sp-CSPL1 algorithm is evaluated on two publicly available EEG data sets, on which it shows significant improvement in classification accuracy.