Energy consumption in personal mobile devices sensing applications

Abstract

Personal mobile devices have a strong impact in the daily life of their users, making part of their daily routines. Most of these devices are equipped with several sensors and interfaces that may be used to study human mobility and its interaction with the elements present in physical spaces. Our goal was to develop an application for Android smartphones that could be used to collect data from several of the devices' interfaces (Wi-Fi, GPS, Bluetooth and GSM), and to send that data to a server for later processing and analysis. In order to maximise the autonomy of the devices, energy resources must be used efficiently. This paper focus on a power-consumption saving solution for mobile phone-based sensing systems in the context of human motion analysis. Experiments were conducted with the objective of comparing power-consumption in different situations using our solution. Results have shown that, considering current power consumption patterns, carefully designed solutions for mobile phone-based sensing for observing human motion may enhance energy efficiency satisfactorily. In this particular domain, we have explored periodic sampling of the sensors and the suspension of the sampling process in the Android operating system whenever the device is not moving and we report such results in this paper.