Increasing automation of buildings enables rich information streams about the activities of building users to reach networked computer systems. Privacy concerns typically cause this information to be accessible only by building managers and security personnel. However, if appropriate privacy mechanisms can be implemented, then it is possible to deploy location information systems that can contribute to the convenience and efficiency of users. This talk describes a three step approach to privacy-sensitive release of location information collected by building sensors. These steps entail defining an ownership model, defining environment events to be monitored, and creating a sharing model. These steps are described mathematically and then validated through a case study for a system called Janus's Map which provides a location information system for the card reader, door, and occupancy sensors of a modern smart building. The talk will conclude with a discussion of future work on this project, including a discussion on the practical aspects of implementing such a system in a real building.
This talk is a practice presentation for a paper of the same title being presented at Security in Pervasive Computing in York, UK April 18-21. The work is joint with Kaijun Tan and Carl A. Gunter.