An adaptive M-wave canceler for the EMG controlled functional electrical stimulator and its FPGA implementation

Biopotential signals have been used as command and feedback signals in systems using electrical stimulation of motor nerves to restore the lost function. In order to use the voluntary electromyography (EMG) as a control signal for the functional electrical stimulation of the same muscle, it is necessary to suppress the M-wave evoked by the electrical stimulation. We present a Gram-Schmidt (GS) prediction error filter which effectively eliminates the M-wave from voluntary EMG. The GS filter has systolic array structure, so it offers advantages for the real-time processing on the field programmable gate array (FPGA). On basis of the data obtained from model for M-wave and voluntary EMG and from CNS injury patient, the proposed GS filter showed a very promising performance.