On reducing energy consumption as a function of space and time in mobile devices

Abstract

Due to the restriction in size and weight of the mobile devices the battery capacity will be limited. There exist a number of factors that affect the consumption of battery, in these devices, such as the running apps, hardware resources such as CPU, GPU, display, mobile network, Wi-Fi, GPS, and Bluetooth radios used by the applications. The non-optimized use of the various system resources will lead to inefficient battery usage in the device. This work studied the average power consumption by the mobile devices as a function of space and time in various scenarios. The variation in power consumption may be contributed by one or more components in different space-time scenarios. One of the main components that causes variation in power consumption is mobile radio. The mobile radio power consumption varies with changes in signal strength and, therefore, energy consumption varies with space and time. The unwanted energy drain in some scenarios can be controlled by carefully identifying spots where high energy consumption can occur and managing the device properly. Based on our study, we propose a classification of the spatio-temporal energy consumption characteristics of the network interface of mobile computing devices into three categories: (i) positively correlated, (ii) negatively correlated, and (iii) neutral, based on the energy consumption and RSSI observations. We identified the scenarios where battery drain becomes high and developed a system to provide alert mechanism that alerts a user to take appropriate actions in such situations so that power can be saved.