Stationary Wavelet-based Two-directional Two-dimensional Principal Component Analysis for EMG Signal Classification

• DOI: 10.1515/msr-2017-0015
• Publication Date: 2017-06-22T10:01:24Z
• Authors: Yi Ji; Shanlin Sun; Hong-Bo Xie
• Keywords: Stationary wavelet transform

Abstract Discrete wavelet transform (WT) followed by principal component analysis (PCA) has been a powerful approach for the of biomedical signals. Wavelet coefficients at various scales and channels were usually transformed into one-dimensional array, causing issues such as curse dimensionality dilemma small sample size problem. In addition, lack time-shift invariance WT can be modeled noise degrades classifier performance. this study, we present stationary wavelet-based two-directional two-dimensional (SW2D 2 PCA) method efficient effective extraction essential feature information from Time-invariant multi-scale matrices are constructed in first step. The then operates on to reduce dimension, rather than vectors conventional PCA. Results presented an experiment classify eight hand motions using 4-channel electromyographic (EMG) signals recorded healthy subjects amputees, which illustrates efficiency effectiveness proposed signal analysis.