ENOrMOUS: ENergy Optimization for MObile plateform using User needS

Abstract Optimizing energy consumption in modern mobile handled devices plays a crucial role as lowering the power consumption impacts battery life and system reliability. Recent mobile platforms have an increasing number of sensors and processing components. Added to the popularity of power-hungry applications, battery life in mobile devices is an important issue. However, the utilization pattern of large amount of data from the various sensors can be beneficial to detect the changing device context, the user needs and the running application requirements in terms of resources. When these information are used properly, an efficient control of power knobs can be implemented to reduce the energy consumption. This paper presents a framework for ENergy Optimization for MObile platform using User needS (ENOsMOUS). This framework is able to identify user contexts and to understand user habits, preferences and needs to improve the operating system power scheme. Machine Learning (ML) algorithms have been used to obtain an efficient trade-off between power consumption reduction opportunities and user satisfaction requirements. ENOrMOUS is a generic solution that manages the power knobs. When applied to the CPU frequency, the sound level, the screen brightness and the Wi-Fi, ENOrMOUS can lower the power consumption by up to 35% compared the out-of-the-box operating system power manager schemes with a negligible overhead.