A Bus-Based Smart Myoelectric Electrode/Amplifier—System Requirements

Feature Extraction and Selection for Myoelectric Control Based on Wearable EMG Sensors

“…Three studies reported that the classification accuracies obtained from a set of multiple time domain features dramatically decrease when the sampling rate is lower than 200 Hz [ 15 ], 400 Hz [ 16 ], or 600 Hz [ 17 ]. In contrast, the other study [ 18 ] reported that reducing the sampling rate does not degrade the classification performance of the similar feature set. This study also examined the performance of myoelectric control as the resolution of analog to digital (A/D) conversion is decreased, and suggested that the resolution of the A/D conversion can be reduced to 8-bit while maintaining or even improving classification performance [ 18 ].…”

“…In contrast, the other study [ 18 ] reported that reducing the sampling rate does not degrade the classification performance of the similar feature set. This study also examined the performance of myoelectric control as the resolution of analog to digital (A/D) conversion is decreased, and suggested that the resolution of the A/D conversion can be reduced to 8-bit while maintaining or even improving classification performance [ 18 ]. Thus, with the growing adoption of lower-bandwidth sensor systems, a better investigation of how the classification performance of EMG feature extraction methods is affected by lower sampling rates is warranted.…”

“…It is clearly seen here that the accuracy in identifying multiple classes of hand and finger motions is shown to benefit from these high frequency components, especially in the case of transradial amputees (a more than difference). Wilson et al [ 18 ] even suggested that removal of the 20–120 Hz frequency band actually increased the classification accuracy of nine classes of motion significantly, and thus they recommend using the combination of sampling rate and high-pass cut-off frequency of 1000 Hz and 120 Hz to maximize classification performance in the presence of power line noise and motion artifacts. The present results agree with and extend upon our preliminary study [ 48 ] (which used a different EMG dataset of 20 able-bodied subjects, and considered another practical robustness issue, i.e., changes in the EMG signal itself over time) as well as those in [ 15 , 16 , 17 ].…”

“…Several studies have investigated the effects of sampling rate on classification accuracy of hand and finger movement tasks down-sampling higher-frequency EMG recordings, but with mixed results [ 15 , 16 , 17 , 18 ]. Three studies reported that the classification accuracies obtained from a set of multiple time domain features dramatically decrease when the sampling rate is lower than 200 Hz [ 15 ], 400 Hz [ 16 ], or 600 Hz [ 17 ].…”

Evaluating Internal Model Strength and Performance of Myoelectric Prosthesis Control Strategies

“…A subject performing the target familiarization block by controlling a cursor on the screen using myoelectric signals sensed by the UNB Smart Electrode System, which is placed on the subjects forearm, to acquire a target movements while providing a comfortable setting during scheduled breaks between testing blocks. A UNB Smart Electrode System [29] was placed on the subject's right forearm ( Fig.2) and the real-time myoelectric signals extracted from muscles contractions were monitored using the Acquisition and Control Environment (ACE) software package [30] developed using MATLAB (Release 2007, The MathWorks, Inc., Natick, Massachusetts, United States).…”