The Influence of Human Walking Activities on the Doppler Characteristics of Non-Stationary Indoor Channel Models

EasyChair Preprint 9221

Abstract

This paper analyzes the time-variant (TV) Doppler power spectral density of a 3D non-stationary fixed-to-fixed indoor channel simulator after feeding it with realistic trajectories of a walking person. The trajectories of the walking person are obtained by simulating a full-body musculoskeletal model in the OpenSim. We provide expressions of the TV Doppler frequencies caused by these trajectories. Then, we present the complex channel gain consisting of fixed scatterers and a cluster of moving scatterers. After that, we use the concept of the spectrogram to analyze the TV Doppler power spectral density of the complex channel gain. Finally, we present expressions of the TV mean Doppler shift and Doppler spread. The work of this paper is important for human activity recognition systems using radio-frequency (non-wearable) sensors as the demand for such systems has increased nowadays.

Keyphrases: OpenSim, Spectrogram, Time-Frequency Distributions, dynamic simulation, musculoskeletal model, non-stationary channel model

@booklet{EasyChair:9221,
  author    = {Muhammad Muaaz and Ahmed Abdel-Gawwad and Matthias Pätzold},
  title     = {The Influence of Human Walking Activities on the Doppler Characteristics of Non-Stationary Indoor Channel Models},
  howpublished = {EasyChair Preprint 9221},
  year      = {EasyChair, 2022}}