Members of the Infospheres group develop theory and then build prototypes to which they apply their theories. The prototype implementation projects are described briefly below.
Radiation Detection
Radiological weapons can cause grievous harm. Explosions of dirty bombs that spray nuclear material may make the public unwilling to enter contaminated areas. The possibility that terrorists may attempt to explode a nuclear bomb over the next 25 years cannot be totally ruled out since mechanisms for doing so are becoming more widely known. Therefore, we are doing research, in collaboration with other groups, on preventing such events. We are developing virtual environments, designing systems with collaborators, and investigating mobile surface and airborne sensors.
Virtual Environments for Radiation Detection
We are developing virtual environments, as layers on top of online games, that represent adversaries carrying nuclear material and first responders attempting to interdict them. Students and law enforcement officials can play roles of both adversaries and security personnel. They can explore the space of strategies available to both attackers and defenders in a safe, inexpensive manner. Different types of sensors, such as those in unmanned aerial vehicles, can be represented within the environment. We model background radiation, threat radiation, photon absorption by walls and other physical aspects of the system. Virtual environments are a useful tool for learning about this complex problem.
Watch demo video on YouTube:
http://www.youtube.com/watch?v=8C0XbRNlouo
Demo video download:
Demo Video 1: 114 MB
Demo Video 2: 132 MB
System Design for Radiation Detection
We are applying S&R theory to the problem of designing systems for interdicting vehicles or pedestrians carrying nuclear material. We are studying a sequence of problems of increasing complexity, starting with detecting single stationary sources, then considering multiple stationary sources, and finally investigating the problem of detecting multiple vehicles, carrying nuclear material, within large geographical regions. We are studying several design alternatives including the relative benefits of sensors with different degrees of directionality, tradeoffs between a few large sensors and several small ones, the efficacy of airborne sensors over open areas, and optimal layout of sensors for buildings and campuses. We plan to make measurements to validate the theory and designs.
The Event Driven Web
An event is a significant change in the state of an enterprise or its environment. An event may be a toxic spill, the detection of radiation material in a container, a candidate winning an election or a steep drop in company stock price. Event-driven applications that are constructed as compositions of Web applications offer significant benefits. Just as “mashups” compose Web services to create added value, so too can compositions of event-driven applications create added value. An event-driven Web detects critical events on the Web and other sources and responds on your behalf; the response often consists of sending a phone or email message or updating an online dashboard. An increasing number of Web sites offer event-driven services. This research investigates the design of a Web of event-driven services.
Personal Monitors
A personal monitor senses events on the Web and elsewhere; it responds by updating a dashboard or by making phone calls or sending messages. Dashboards for business activity monitoring (BAM) are becoming increasingly used in business, and personal dashboards are becoming common. This research investigates tools for the development of personal monitors dealing with a variety of interests including politics, businesses, science, sports and entertainment. The theoretical aspects of this work deal with specifications and detection of events. The prototype implementations deal with displaying results from monitoring in the most convenient ways for end users.