Abstract
Overall high power consumption in the mobile applications forces the mobile users to recharge frequently. Most of the Android applications do not implement any self-adaptive strategies that react to the battery level, status and context. Thus the applications continue to consume power even when battery is critically low. Intelligent control of hardware and software optimization based on the battery level is the key to power saving. This paper introduces a self-adaptive application development framework which proposes three profiles with various self-adaptive features for mobile applications. The framework employs an analyzer engine which decides the activation of appropriate profile based on battery and context information. The self-adaption takes place in four levels - hardware & software features adaption, user features adaption and additional optimization. When the battery is critically low, priority is given to maximize the battery life until next charging opportunity. Such implementation is highly desirable for mobile applications with high dependency on display hardware (e.g. games) and/or on network operations (e.g. YouTube, Dropbox). Prototype Android applications are developed and results show up to 40% reduction in application power consumption. Power Tutor has been used to get the power consumption results.