Journal of Applied Statistics · 2015 · 40 citations · 16 references
EngineeringBiomedical EngineeringBiomechanicsMotor ControlMovement AnalysisElectromyogram AnalysisMovement Pattern AnalysisMuscle Signal ProcessingElectromyographyElectromyography Feature ExtractionElectromyography Pattern RecognitionElectromyography Signal ClassificationMultiresolution AnalysisMuscle Signal Feature ExtractionShort-time Fourier TransformWavelet Transform
In this investigation, extracted features ofsignals have been analyzed for the recognition of arm movements. Short-time Fourier transform and wavelet transform based on Euclidian distance were applied to reordered signals. Results show that wavelet is a more useful and powerful tool for analyzing signals, since it shows multiresolution property with a significant reduction in the computation time for eliminating resolution problems. Finally, a statistical technique of repeated factorial analysis of variance for experimental recorded data was implemented in a way to investigate the effect of class separability for multiple motions for establishing surface electromyogram–muscular force relationship.
16