Lightweight energy consumption analysis and prediction for Android applications

Abstract

The energy consumption problem is a hot topic in Android communities. The high energy cost caused by improper development brings lots of complaints from users. An effective and efficient energy consumption analysis technique can guide Android developers to improve the energy efficiency of their apps. Existing researches on this problem focus on either system entity level that gives the energy consumption of the hardware, or source line level that calculates the energy cost of source codes. With the consideration of accuracy and cost of analysis, this paper proposes a lightweight and automatic approach to analyze and predict the energy consumption for Android apps. We conduct the study from a method-level and API-level perspective. The method-level analysis gives developers facts about the energy consumption of the user methods in their apps, while the API-level analysis shows the energy consumption of Android APIs, which can help them make good decisions about how to choose appropriate APIs to improve the energy efficiency of an Android app. We construct a statistical model from a set of energy values obtained by Dalvik bytecode based instrumentation and software-based measurement, to predict the energy consumption of method sequences or API sequences. The experiments on several real-world apps show that the proposed techniques have low overhead while persisting acceptable accuracy. • An automatic approach to construct accurate energy model for Android app is proposed. • An optimized linear regression algorithm to induce energy model. • API level energy prediction is studied, modeling energy consumption of Android APIs.