Output details
15 - General Engineering
University of Kent
ParkSense: A Smartphone-Based Sensing System For On-Street Parking
The paper presents a novel technique for a smartphone-based sensing system that detects parking behaviour, while avoiding the capture of actual location traces of the user. Parking and driving detection relies on a robust WiFi signature matching technique that maintains a low energy footprint, consuming 26% of the energy consumed by a network-based approach, and 48% of a GPS-based approach. It is published at the highly respected ACM MobiCom Conference. Following this work, a collaboration with Parkopedia (Eugene Tsyrklevic eugene@parkopedia.com) was established offering us access to data and infrastructure resources, in order to design an integrated parking prediction system.