Power Based Malicious Code Detection Techniques for Smartphones

Abstract

In recent years there have been a growing number of viruses, root kits, and malware designed to gain access to system resources and information stored on smartphones. This paper examines the effectiveness of two smartphone-specific techniques for detecting malicious code behavior based on individual power consumption profiles based on time and location. Prototypes for each of the techniques were developed for use with Google's Android platform and were each evaluated over periods of six months or more involving several real-world users. This paper describes the design and implementation of each of these prototypes as well as the evaluation of their results.