Applications of electrical stimulation for restoration of functional movements such as standing, gait, and grasp have always been hindered by the rapid fatigue of stimulated muscle. This paper describes an experimental investigation of stimulation with N-lets (a set of N closely spaced stimulation pulses) as a means of producing contractions with improved fatigue characteristics. Experiments were conducted on 27 able-bodied and four SCI human subjects using surface stimulation of the quadriceps muscle to produce isometric knee joint torque. Based upon evidence from the literature on muscle fatigue, parameters of the N-let trains for N = 1-6 were optimized to produce the most force per pulse. The results demonstrated that: 1) nonlinear summation of the twitch response occurs in human subjects with N-let surface stimulation; 2) for most subjects, doublet stimulation (N = 2) with a pulse interval of about 5 ms produced the maximum torque-time integral per pulse of the resulting twitch; and 3) on average, optimal N-let stimulation resulted in a 36% increase in isometric torque tracking when compared to traditional singlet stimulation. The results have immediate implications for alleviating the problem of premature fatigue during functional electrical stimulation.

This work reports on a method to measure the instrumental noise of a sensor in the presence of large background signals, based on measuring the coherence and output of two identical sensors exposed to the same input stimulus. The technique can be applied to any sensor characterization problem where there is interest in distinguishing random noise from common-mode noise. The method is demonstrated using geophones, a single-axis seismometer that has been an important sensor for geophysical measurements for decades. Previous experimental work has not published measurements of the geophone’s instrumental noise at frequencies near the geophone’s resonance because of the presence of large background seismic signals compared to the predicted instrumental noise, which is a minimum at the geophone’s resonant frequency. With this coherence technique, measurements performed in the presence of 1 μg/Hz seismic signals showed that the minimum instrumental noise was within a factor of 2 of the predicted minimum noise of 0.6 ng/Hz.

No abstract

No abstract

No abstract