A novel signal processing method for multi-electrode surface electromyography

The signal to noise ratio (SNR) of respiratory electromyography captured by surface electrodes is very low. The EMG signal is disturbed by many types of noises; especially the cardiac artifact (ECG). The amplitude ratio between the cardiac artifact and the EMG signal is variable because of the non-stationary nature of the EMG signal and this makes the separation complex. The authors suggest here a new signal processing method based on a finite impulse response (FIR) adaptive filter. The proposed solution assumes that the acquisition is carried out by using multi-electrode array. The idea is to have a good estimation of the ECG artifact in order to use it as a reference noise signal for the authors´ adaptive filter. However, the final algorithm was executed in real-time using a DSP-PC board, where the EMG signal was acquired by a surface electrode array