Mobility Mode Detection Using WiFi Signals

TL;DR

This paper presents a method to classify mobility modes such as walking, biking, and driving using Wi-Fi signals and machine learning, achieving high accuracy in a real urban environment.

Contribution

It introduces a novel approach combining Wi-Fi signal analysis with deep learning and decision trees for mobility mode detection in urban settings.

Findings

Multilayer Perceptron achieved 86.52% accuracy.

Wi-Fi signals can effectively distinguish mobility modes.

Decision tree classifiers performed slightly worse than neural networks.

Abstract

We utilize Wi-Fi communications from smartphones to predict their mobility mode, i.e. walking, biking and driving. Wi-Fi sensors were deployed at four strategic locations in a closed loop on streets in downtown Toronto. Deep neural network (Multilayer Perceptron) along with three decision tree based classifiers (Decision Tree, Bagged Decision Tree and Random Forest) are developed. Results show that the best prediction accuracy is achieved by Multilayer Perceptron, with 86.52% correct predictions of mobility modes.