Power Consumption Profiling Method Based on Android Application Usage

Abstract

In this paper, we propose a method for collecting essential data to profile energy consumption of applications running on Android OS. Existing power-estimation methods are unable to account for all possible usage patterns, since developers can only prepare a limited number of profiling test cases. Our proposed method analyzes the power consumption using a log collected during an application use on a smart phone of a particular user. In our method, the logging code that tracks application usage data for the user is automatically embedded into the application. Our method uses this usage information to estimate power consumption, and provide developers with helpful hints for decision making and application tuning. In this paper, we analyzed the power consumption of an open-source application based on the log collected and estimated the power-saving effects of the application. The power consumption of the application was reduced by tuning the application according to the data derived from the results of the analyses; thus confirming that the method provides valid information to determine power-saving techniques.