Enabling Massive Scale Sensing with the @LooCI Mobile Sensing Framework

Abstract

21st century objectives such as smart cities require sensing at massive scale. To cost effectively realize massive-scale sensor networks, we advocate for the integration of purpose-deployed sensor networks with consumer mobile devices such as smart phones and web pads. These mobile devices serve as a local management interface to the sensor network, or as sensor nodes themselves in a participatory sensing model. To realize this vision, a common middleware is required that allows heterogeneous nodes with different hardware resources, network media and software stacks to be composed into coherent distributed applications. This paper extends the LooCI sensor network middleware with a framework for supporting mobile consumer devices. The resulting middleware, which we call @LooCI, provides a consistent development and management environment for both embedded sensor nodes and consumer mobile devices. @LooCI uses social network technologies to manage mobility and to facilitate the composition of sensors with human actors. We report on a prototype implementation of @LooCI for Android and Twitter. Our evaluation shows that @LooCI provides good performance on mobile devices and that social networks are a promising mechanism to integrate sensors with consumer mobile devices.