Funded by:
(grant reference EP/D076625/2)
Dates: October 2006 - September 2010
Main project webpage: http://www.ubival.org/
Institutions:
University of Oxford
University of Birmingham
University College London
Imperial College London
Industrial Project Partners:
British Telecommunications Plc (BT)
Hewlett-Packard Ltd
QinetiQ (Malvern)
Mark Weiser's vision of ubiquitous computing, in which computers become transparently and seamlessly woven into the many activities of our daily lives, is slowly becoming a reality. Researchers have created prototype ubiquitous computing environments such as 'smart homes' that can automatically sense the presence of a resident in a particular room and change some aspect of the environment of the room such as turning on the lights, or 'smart museums' that can play recorded information about the museum artefact a visitor is standing in front of. There seem to be limitless possibilities for the kinds of environments and applications that can be developed for ubiquitous computing, yet the very nature of ubiquitous computing creates new and significant challenges for engineers who would like to build these environments and applications. Anybody who has ever used a computer has experienced the extreme frustration of using a software package that doesn't work the way it's supposed to, or that unceremoniously crashes in the middle of its operation, or that runs extremely slowly, or that transmits sensitive information such as credit card numbers over untrusted networks. For ubiquitous computing to achieve true transparent and seamless integration with its surroundings, it is important to prevent such mishaps, crashes, inefficiencies and insecurities from happening to the greatest extent possible. This project will define and implement a suite of sound, systematic methods that engineers can use to create correctly functioning, efficient and secure ubiquitous computing environments and applications. The research will be conducted and evaluated using the smart urban spaces and applications being developed in another ubiquitous computing project called Cityware.
This project has defined and implemented a suite of sound, systematic methods that engineers can use to create correctly functioning, efficient and secure ubiquitous computing environments and applications. At Oxford and Birmingham, the research focused on formulating probabilistic model checking techniques for analysing performance and reliability of ubiquitous computing applications; designing a framework for simulation and verification of wireless network protocols, including multi-objective specifications; analysing network security, anonymity and access control protocols; and developing software verification techniques and tools for sensor networks. The research was conducted and evaluated using several significant case studies, e.g. the Zigbee protocol, for which a range of performance and energy measures were automatically calculated for a variety of scenarios to identify possible improvements the design.
Publications: