Multi-Scale Analysis of Myoelectric Signals: Assessment of Long-Range Dependencies and Fractal-Scaling-Break

Abstract

In this paper we propose a novel methodology for structure-function-based multi-fractal analysis of myoelectric signals. This methodology provides multi-scale information about the geometry of the myoelectric interference pattern. Specifically, it provides insight into the fractal characteristics of sampled myoelectric signals with assessment of long-range dependencies and fractal-scaling-break properties of this signal. Power spectrum and structure-function-based methods are also integrated in this work, presenting a unified framework for multi-scale analysis of myoelectric signals. Results of an experiment for comparison of myoelectric signals to strict mathematical fractional Brownian motion are provided. The novel methodology provides insight into the myoelectric signal’s renewal process. The results also show a great potential for applications to clinical diagnosis and fatigue studies.