An optimal real-time digital processor for the electric activity of muscle

The electric activity of muscle generates a biologic signal that can be used to estimate muscle force. However, large-amplitude, low-frequency variations are frequently superimposed on estimates of muscle force that are based on myoelectric signals. Conventional myoelectric signal processors reduce this "noise" component by simple filtering, which adversely affects the speed of response of the force estimate. Based on a novel algorithm for processing raw myoelectric signals that optimally reduce these random fluctuations, a new myoelectric signal processor was developed with hardware-assisted digital microprocessor technology; two sets of four differential electrodes are sampled at a rate of 1 kHz to generate two independent estimates of muscle force. Use of this processor resulted in an almost fivefold improvement in precision of force estimate compared with estimates obtained from the conventional myoelectric processor. If the new device were reprogrammed slightly, the speed of response achieved could be as much as 20 times greater than that of the conventional analog processor.