Smartphone Impostor Detection with Behavioral Data Privacy and Minimalist Hardware Support

TL;DR

This paper introduces a privacy-preserving, real-time impostor detection system for smartphones using a lightweight RNN-based deep learning approach that leverages behavioral sensor data and minimal hardware support.

Contribution

It presents a novel RNN-based algorithm that protects user privacy by not exposing sensor data outside the device and integrates a low-cost hardware module for real-time detection.

Findings

SID supports real-time impostor detection

Achieves higher accuracy than prior machine learning methods

Low hardware cost and energy consumption

Abstract

Impostors are attackers who take over a smartphone and gain access to the legitimate user's confidential and private information. This paper proposes a defense-in-depth mechanism to detect impostors quickly with simple Deep Learning algorithms, which can achieve better detection accuracy than the best prior work which used Machine Learning algorithms requiring computation of multiple features. Different from previous work, we then consider protecting the privacy of a user's behavioral (sensor) data by not exposing it outside the smartphone. For this scenario, we propose a Recurrent Neural Network (RNN) based Deep Learning algorithm that uses only the legitimate user's sensor data to learn his/her normal behavior. We propose to use Prediction Error Distribution (PED) to enhance the detection accuracy. We also show how a minimalist hardware module, dubbed SID for Smartphone Impostor Detector, can be designed and integrated into smartphones for self-contained impostor detection. Experimental results show that SID can support real-time impostor detection, at a very low hardware cost and energy consumption, compared to other RNN accelerators.