PowerSpy: Location Tracking using Mobile Device Power Analysis

TL;DR

This paper demonstrates that aggregate power consumption data from mobile devices can be exploited to infer user location, revealing a privacy vulnerability despite the data's noisy nature.

Contribution

It introduces a novel method using machine learning to infer user location from aggregate power data without requiring permissions.

Findings

Power data can reveal user location with high accuracy.

Machine learning effectively filters noise in power consumption data.

Privacy risks exist even when data is considered harmless.

Abstract

Modern mobile platforms like Android enable applications to read aggregate power usage on the phone. This information is considered harmless and reading it requires no user permission or notification. We show that by simply reading the phone's aggregate power consumption over a period of a few minutes an application can learn information about the user's location. Aggregate phone power consumption data is extremely noisy due to the multitude of components and applications that simultaneously consume power. Nevertheless, by using machine learning algorithms we are able to successfully infer the phone's location. We discuss several ways in which this privacy leak can be remedied.