Smart Urban-Scale Sensing

Cities are becoming increasingly instrumented as the number and types sensors being deployed increases. This offers the opportunity for more fine-grained information about the patterns of use of city resources and systems to be gathered. The resulting information could be used by city authorities to optimize delivery of services. Equally, such information when made available to the public at large can contribute to e-deliberation and e-democracy as citizens discover how their shared resources are being used and can influence public representatives and decision-makers by analyzing the data that is collected.

Scientific Basis

Wireless sensor networks (WSNs) are now moving out of the laboratory and being deployed in the real world on a large scale. Unlike traditional sensor networks, a city’s sensor infrastructure does not just consist of installed sensor networks (e.g., traffic detectors, smart energy or water meters) deployed and maintained by city authorities and public utilities. Millions of sensor-enabled smart phones being carried by its citizens are a vital new source of data as are the thousands of vehicles equipped with sensory devices such as in-vehicle positioning or diagnostic systems as well as the hundreds of private closed-circuit cameras facing our streets and car parks. These sensors vary widely in terms of what they sense, how often they sense, whether and how often they transmit the data they collect, the means of communication they use, what their source of power is and who owns them. Smart-scale urban sensing thus tackles the scientific challenges associated with gathering and analyzing city-wide sensor data from a heterogeneity of sources.

Research Areas

In particular, we are investigating the following areas:

Middleware for Urban-Scale WSNs: the design of middleware that supports in-network data fusion and aggregation to handle a huge number of sensors,
Unstable Path Routing in WSNs: communication algorithms that can exploit temporary and unstable paths effectively to send data in the most energy efficient manner,
Energy Conservation in WSNs: intelligent mechanisms that enable only required sensors to fulfil application requirements to sense and transmit, saving energy and reducing unnecessary communication traffic,
Participatory Sensing: investigating the practical problems, in terms of data accuracy in particular, encountered when deploying an application that relies user-contributed sensor data
Development Environments for Sensor Network Applications (DESNA): designing an integrated tool chain for sensor network applications.