Signal Processing in Movement Analysis (a state-space approach)

Signalprocessing techniques in movementanalysis (MA) are a pre-requisite for further processing of MA data and can heavily affect the reliability of these latter results. In this paper, a brief review of classical digital signalprocessing techniques used in MA will be outlined, but emphasis will be given to a particular class of methods based on the state-spaceapproach and in particular on the application of Kalman filtering theory. It will be shown the use of these techniques for the solution of linear and non-linear filtering problems such as those, respectively, relative to the numerical smoothing/differentiation of noisy signals and relative to the filtering of multiple displacement data subject to kinematic constraints.