A Personalized Profile Based Learning System For Power Management In Android
MetadataShow full item record
Mobile computing devices are becoming more ubiquitous everyday due to the phenomenal growth in technology powering them. With the amount of computing power available in these devices, users are capable of achieving a multitude of tasks that were only possible with a PC just a few years ago. However, these devices still face issues regarding power management. Battery technology has not kept pace with the development in other areas. With a limited supply of energy, the mobile device of today requires a fine balance of power management to provide adequate energy to support the heavy duty computing of the user while simultaneously enabling the device to stay alive for a long duration. With such limitations, the onus is more on the user to limit his usage of the device and its features to conserve power, thus having a crippling effect on the user's operation of devices. This research effort analyzes how devices are used and explores the effect of demographics on power consumption. We also propose a solution which will adapt to the individual user and provide a customized power saving mechanism tailored to the user's usage of his/her device. The adaptive system will learn what type of apps are used by the user and can intelligently make decisions to conserve power based on prior learnings. It is estimated that such a mechanism will have an improvement on battery life by 15%.