Haze: Privacy-Preserving Real-Time Traffic Statistics

Abstract

We consider traffic-update mobile applications that let users learn traffic conditions based on reports from other users. These applications are becoming increasingly popular (e.g., Waze reported 30 million users in 2013) since they aggregate real-time road traffic updates from actual users traveling on the roads. However, the providers of these mobile services have access to such sensitive information as timestamped locations and movements of its users. In this paper, we describe Haze, a protocol for traffic-update applications that supports the creation of traffic statistics from user reports while protecting the privacy of the users. Haze relies on a small subset of users to jointly aggregate encrypted speed and alert data and report the result to the service provider. We use jury-voting protocols based on threshold cryptosystem and differential privacy techniques to hide user data from anyone participating in the protocol while allowing only aggregate information to be extracted and sent to the service provider. We show that Haze is effective in practice by developing a prototype implementation and performing experiments on a real-world dataset of car trajectories.