Spheres: A Web Services Framework for Smartphone Sensing as a Service
Abstract

Smartphones are mobile phones that execute application programs. Smartphones are practically everywhere, in every country and culture. Smartphones can tell where they are by GPS, a spatial sense, and what time it is, a temporal sense, relative to their GPS location. Groups of smartphone users form communities, creating a social sense. Smartphones often contain built-in sensors including temperature, proximity, accelerometer, gyroscope, magnetometer, barometric pressure, ambient light, camera, microphone, etc., as well as the ability to connect to external sensors over their wireless interfaces forming wireless sensor networks. Smartphones are connected to the Internet through their cellular data connections providing a globally established deployment of “things” on the Internet of Things (IoT). People carrying their smartphones represent a global source of continuous mobile Internet-connected sensor data. The resulting human, social, and environmental data represents an untapped global information resource of epic proportion. This paper explores the current research related to smartphone sensing as a service (S2aaS) and proposes an architecture for a web services framework for S2aaS called Spheres. Spheres will integrate wireless sensor network, mobile Internet, and cloud technologies into an open-source service-oriented architecture (SOA) providing crowdsourced smartphone sensor data aggregation and sharing in secure cloud deployments for private and public consumption. Spheres collected data can be overlaid visually on maps or incorporated into applications that require the data from sensors embedded in, or connected to, smartphones over specific regions and times of interest. In addition to specifying an architecture, this paper will identify the challenges expected to implement and demonstrate a prototype of Spheres on at least one smartphone platform with the cloud platform and web services deployed on the UMBC BlueWave high performance computing system.