A framework for usage pattern–based power optimization and battery lifetime prediction in smartphones

Abstract

The use of mobile devices has increased many folds over the last few years. Smart phones are used not only for communication but also for storage of personal contents like photos, videos, documents, and bank account and credit/debit card details. Secure access to these contents is very important. Nowadays, many mobile devices come with inbuilt biometric-enabled security features like fingerprint, face recognition, and iris. However, additional battery power is consumed each time a user unlocks the device using any one of these security features. In order to enable prolonged use of the device, there is a strong need to find ways to conserve power in mobile devices. At the same time, it is also equally important that the smart phone user knows how long the battery of his device will last. In this paper, we present a novel power optimization and battery lifetime prediction framework called P4O (Pattern, Profiling, Prediction, and Power Optimization). Our contributions are threefold—(i) Propose a novel framework for power optimization in smart phones. (ii) Propose a new approach for battery lifetime forecast. (iii) Implement and validate the efficacy of the proposed framework. For experimental results, the proposed framework was implemented on the Android-based smartphone. The experimental results validate the proposed framework with power optimization up to 40% over default Linux and Android power saving features available in an Android operating system. This framework is also able to forecast battery lifetime with accuracy of up to 98%.