Evaluating the iPhone as a Mobile Platform for People-Centric Sensing Applications

A number of mobile phones such as the Nokia N95 and Apple iPhone are being used by researchers to explore new people-centric sensing applications. These top-end phones include various sensors (e.g., accelerometer, proximity sensor, GPS, camera, microphone), radios (e.g., Bluetooth, WiFi, cellular), operating systems (e.g., Symbian, customized Mac OS X), and processors (e.g., 330 Mhz ARM, 412 Mhz ARM). While there is some data on the Nokia N95, little, however, is known about the ability of Apple’s iPhone to support the necessary sensing, processing and communications needs of these emerging sensing applications. We present the first quantitative performance evaluation of the iPhone’s sensors, localization engine, and networking stack while running CenceMe, a representative people-centric sensing application. We profile the performance of CenceMe running on the iPhone in terms of energy consumption and computational speed of its algorithms. One drawback of using the iPhone over the N95 is that it does not allow third party applications such as CenceMe to run as a background process, making continuous sensing problematic. The upside is that the iPhone offers a rich UI architecture, high computational capability, and an efficient application distribution system through Apple’s App Store.