Novel method for stride length estimation with body area network accelerometers

Gait analysis using wireless accelerometers deployed as body area networks can provide valuable information for multiple health-related applications. Within this field, stride length estimation represents a difficult task. In this paper we present a novel method to estimate stride length through the application of the wavelet transform to the signal obtained from a wireless accelerometer on the waist. We also introduce a novel metric to determine the level of the wavelet transform detail coefficients from which the step frequency can be directly extracted. Additionally, we show the correlation between the energy of the wavelet transform approximation coefficients and the speed of the gait.

I. Introduction

Gait analysis through wireless accelerometers deployed as body area networks can provide useful information for multiple health-related applications, including: assessment of physical activity, diagnosis of medical conditions and assistance in planning pharmacotherapy and rehabilitation for different patients. The analysis of spatio-temporal parameters of human walking such as displacements at each swing or total stride length, is a complex and difficult task. Multiple studies use sensors attached to different parts of the body including waist, thigh, shank and foot [1]–[3], with special focus on the estimation of temporal features of gait including stride time and cadence. Pedestrian dead reckoning systems make use of accelerometers for step counting and stride length estimation, together with magnetic compasses or gyroscopes to obtain directions. The main errors of this approach relate to the stride length calculation, degrading the accuracy of the solution over time, with average errors of known systems ranging from 3% to 15% of the distance [4], [5].