Surface electromyography (sEMG) is used to study underlying muscle activity. However, due to anatomical complexity, there is cross-talk, and the recorded sEMG corresponds to the mix of activities of different muscles. To overcome this, studies have attempted the use of blind source separation (BSS) and also the use of arrays of electrodes to identify activity of individual muscles. However, it is difficult to determine the most appropriate set of electrodes due to the dependency between the various channels. This paper describes a method by which it is possible to identify the dependency between the various channels, and thus the most appropriate set of channels can be identified. This study has investigated different configurations of electrodes for accurate identification of different hand gestures required for human-computer interface application.
- Bio-signal analysis
- Blind source separation (BSS)
- Hand gesture sensing
- Independent component analysis (ICA)
- Surface electromyography (sEMG)