Modeling of Noisy Acceleration Signals From Quasi-Periodic Movements for Drift-Free Position Estimation

We present a novel approach to drift-free position estimation from noisy acceleration signals, which often arise from quasi-periodic small-amplitude body movements. In contrast to the existing methods, this data-driven strategy is designed to properly describe time-variant harmonic structures in single-channel acceleration signals for low signal-to-noise ratios.

It comprises three processing steps: 1) short-time modeling of acceleration dynamics (instantaneous harmonic amplitudes and phases) in the analysis frame, 2) analytical integration that yields short-time position, and 3) overlap-add recombination for full-length position synthesis.

The comparative results, obtained from the medio-lateral X-acceleration components from 30-s chair stand test recordings, suggest that the proposed method outperforms two state-of-the-art reference methods in terms of Euclidean error, root mean square error, correlation coefficient, and harmonic-to-noise ratio.

A major benefit of the method is that acceleration signal components unrelated to movement are suppressed in the whole analysis bandwidth, which allows for position estimation completely free of low-frequency artifacts.

We believe that the method can be useful in frailty assessment in elderly population, as well as in clinical applications related to gait analysis in aging and rehabilitation.