Inference attack in Android Activity based on program fingerprint

doi:10.1016/j.jnca.2018.12.007

Return

Inference attack in Android Activity based on program fingerprint

2019-02-01

PRE

DOI：10.1016/j.jnca.2018.12.007

Original

Original request for help

Save

Abstract

En 中文

Private breach has always been an important threat to mobile security. Recent studies show that an attacker can infer users' private information through side channels, such as the use of runtime memory and network usage. For side-channel attacks, malicious applications generally run parallel in the background with a foreground application and stealthily collect side-channel information. In this paper, we analyze the relationship between memory changes and Activity transition, then use side-channel information to label an Activity and build an Activity signature database. We show how to use the runtime memory exposure to infer the Activity transition of the current application and use other side channels to infer its Activity interface. We demonstrate the effectiveness of the attacks with 5 popular applications that contain user sensitive information, and successfully inferred most of the Activity transition and Activity interface process. Moreover, we propose a protection scheme which can effectively resist side-channel attacks.

Keywords:

Android security

Side-channel attack

Activity inference

Program fingerprint

Protection

AI Summary

Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.