Technical Report NWU-EECS-10-06

We present a comprehensive analysis of real smartphone usage during a 6-month study, involving 25 users, and one specific smartphone, the Android G1. The goal of our work is to study the high-level characteristics of smartphone usage, and to understand the implications on optimizing smartphones, and their networks. We present 11 findings that cover general usage behavior, interaction with the battery, power consumption, network activity, frequently-run applications, and modeling usage states. We then discuss the implications of these findings on optimizing and/or modeling smartphones and their networks. Some of our findings include the following: (1) we show that battery management is a significant part of the user experience, with near daily charging and frequent low battery indicators; (2) we characterize the system-level power consumption to show that the screen and CPU are two of the most power hungry components on a smartphone; (3) we characterize EDGE and wifi session, providing insight into developing models for smartphone networks; and (4) we demonstrate that real user activity can be clustered to yield a small number of machine states, and state transitions, that can form the basis for modeling smartphone devices.