Activity Recognition from Smart Insole Sensor Data Using a Circular Dilated CNN

TL;DR

An activity classification system based on a circular dilated convolutional neural network that processes multi-modal time-series data from smart insoles that reveals that inertial sensors (accelerometer and gyroscope) contribute substantially to discrimination.

Abstract

Smart insoles equipped with pressure sensors, accelerometers, and gyroscopes offer a non-intrusive means of monitoring human gait and posture. We present an activity classification system based on a circular dilated convolutional neural network (CDCNN) that processes multi-modal time-series data from such insoles. The model operates on 160-frame windows with 24 channels (18 pressure, 3 accelerometer, 3 gyroscope axes), achieving 86.42% test accuracy in a subject-independent evaluation on a four-class task (Standing, Walking, Sitting, Tandem), compared with 87.83% for an extreme gradient-boosted tree (XGBoost) model trained on flattened data. Permutation feature importance reveals that inertial sensors (accelerometer and gyroscope) contribute substantially to discrimination. The approach is suitable for embedded deployment and real-time inference.

Figures (5)

• Figure 1: Class distribution of the compiled smart insole dataset.
• Figure 2: Example tri-axial accelerometer window for a single activity segment.
• Figure 3: Confusion matrix for CDCNN model on the held-out test set.
• Figure 4: Permutation feature importance by channel over the test set.
• Figure 5: Permutation feature importance by channel over the test set for the XGBoost.