Peter F Lamb1, Michael Stöckl2. 1. School of Physical Education, Sport and Exercise Sciences, University of Otago, 56 Union St West, Dunedin 9016, New Zealand; Faculty of Sports and Health Science, Technische Universität München, Georg-Brauchle-Ring 60/62, 80336 Munich, Germany. Electronic address: peter.lamb@otago.ac.nz. 2. Department of Biomechanics, Kinesiology and Applied Computer Science, University of Vienna, Auf der Schmelz 6A, A-1150 Vienna, Austria.
Abstract
BACKGROUND: In this paper we review applications of continuous relative phase and commonly reported methods for calculating the phase angle. Signals with known properties as well as empirical data were used to compare methods for calculating the phase angle. FINDINGS: Our results suggest that the most valid, robust and intuitive results are obtained from the following steps: 1) centering the amplitude of the original signals around zero, 2) creating analytic signals from the original signals using the Hilbert transform, 3) calculating the phase angle using the analytic signal and 4) calculating the continuous relative phase. INTERPRETATIONS: The resulting continuous relative phase values are free of frequency artifacts, a problem associated with most normalization techniques, and the interpretation remains intuitive. We propose these methods for future research using continuous relative phase in studies and analyses of human movement coordination.
BACKGROUND: In this paper we review applications of continuous relative phase and commonly reported methods for calculating the phase angle. Signals with known properties as well as empirical data were used to compare methods for calculating the phase angle. FINDINGS: Our results suggest that the most valid, robust and intuitive results are obtained from the following steps: 1) centering the amplitude of the original signals around zero, 2) creating analytic signals from the original signals using the Hilbert transform, 3) calculating the phase angle using the analytic signal and 4) calculating the continuous relative phase. INTERPRETATIONS: The resulting continuous relative phase values are free of frequency artifacts, a problem associated with most normalization techniques, and the interpretation remains intuitive. We propose these methods for future research using continuous relative phase in studies and analyses of humanmovement coordination.
Authors: Ana F Silva; Pedro Figueiredo; João P Vilas-Boas; Ricardo J Fernandes; Ludovic Seifert Journal: Int J Environ Res Public Health Date: 2022-06-08 Impact factor: 4.614