A Π-Σ network based EMG identification method

Author

Zhang, Haihong ; Cai, Liyu ; Wang, Zhizhong

Author_Institution

Dept. of Biomed. Eng., Shanghai Jiaotong Univ., China

Volume

1

fYear

1999

fDate

1999

Abstract

It is very important for the development of prosthesis to improve the technique of automatic pattern recognition of EMG signals. The conventional means have the disadvantage of slow training speed and limitation in problem scale. This paper proposes a new method which adopts a novel feedforward network called the Π-Σ network to identify the EMG features extracted through wavelet transformation. Experiments results show good convergence properties, high identification rate and especially faster learning speed compared with traditional methods

Keywords

convergence of numerical methods; electromyography; feature extraction; feedforward neural nets; medical signal processing; pattern classification; signal classification; wavelet transforms; Π-Σ network; EMG signal identification method; automatic pattern recognition; convergence properties; faster learning speed; feedforward network; high identification rate; high-order network; mean square error; nonlinear features discrimination; prosthesis development; surface electrode signals; wavelet transform; Electromyography; Function approximation; Interference; Multilayer perceptrons; Neural networks; Neural prosthesis; Pattern classification; Signal processing; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

[Engineering in Medicine and Biology, 1999. 21st Annual Conference and the 1999 Annual Fall Meetring of the Biomedical Engineering Society] BMES/EMBS Conference, 1999. Proceedings of the First Joint

Conference_Location

Atlanta, GA

ISSN

1094-687X

Print_ISBN

0-7803-5674-8

Type

conf

DOI

10.1109/IEMBS.1999.802644

Filename

802644