"There is currently no such competition in the wireless networking field," said Luiz DaSilva, associate professor. "We were inspired by open competitions in the areas of robotics, software engineering, and automotive design that are successful in energizing students to perform hands-on work while advancing the state of research."
"These competitions are very motivating, not to mention fun. Also, failure often teaches us more than success, and implementation is always more convincing than simulation," DaSilva said. "The kind of informal exchange of ideas that occurs naturally in a competition like this tends to move research forward in unexpected ways."
With the advent of wireless ad hoc networks, some of the toughest issues in communications networking are questions of cooperation: Will users trade off bandwidth, signal strength, or speed to ensure system effectiveness? If so, how? What are the incentives that will get users to provide services (such as routing) to other nodes?
DaSilva and Allen MacKenzie, ECE assistant professor, hope to get answers through this wireless networking contest they are developing. It will further networking techniques and algorithms, and provide one-of-a-kind opportunities to study actual, uncontrolled, ad hoc networks, where users make their own decisions regarding tradeoffs between self-interest and common network goals.
The competition will be held for two consecutive years, 2007 and 2008, in conjunction with a major networking conference, yet to be announced. The entries will be judged on end-to-end performance, energy efficiency, and engineering design.
Teams competing in the MANIAC Challenge will match their algorithmic prowess in sending data across uncontrolled, ad hoc networks, where users employ different hardware and software. They will be judged on speed and efficiency. The 2007 competition will entail a video and data relay race in which all the teams are on the mobile ad hoc network (MANET), but no team will be able to have their traffic delivered without cooperation from others.
The software developed at Virginia Tech will also monitor node behavior and system effectiveness, giving researchers a goldmine of data on actual ad hoc network behavior in a field dominated by simulation and controlled testbed research. "There has recently been some soul searching by the networking community regarding the prevailing use of simulation as the main research methodology," explained MacKenzie. "Questions linger about how well MANETS will work in the wild -- outside of tightly controlled lab environments or military deployments. Are simulation results reported in the literature too optimistic about performance that can be achieved in these networks? This competition will provide researchers with a unique opportunity to study real-life network behavior in the wild."
In studying the MANIAC Challenge, researchers will be using the experimental method, which is more common in the natural and social sciences than in engineering. "In this sense, the experiments will be similar to some conducted in sociology and economics, where human behavior is observed in an uncontrolled environment," said MacKenzie. "Increasingly, we are seeing ties between research questions in networking and those in the world of behavioral sciences."
The two-in-one effort is aimed at meeting educational and research goals of improving network throughput, deepening understanding of overall network behavior, and motivating students in the field.