WIRELESS – This Game’s On The Air

A Beautiful Mind, Ron Howard’s Oscar-winning film about noted economist John Nash, briefly gave popular regard to economic game theory—that is, the idea that relatively simple rules can be woven together to create models of complex economic relationships. But game theory may have additional relevance beyond economics and Academy Award nominations. Allen MacKenzie, an assistant professor in the department of electrical and computer engineering at the Virginia Polytechnic Institute and State University (a.k.a. Virginia Tech), is trying to apply game theory to another complex and seemingly amorphous pursuit: wireless networking.

“I think there’s a recognition in the wireless networking community that we’re missing some of the tools we need to really understand how wireless networks work,” MacKenzie says. The current state of wireless network analysis, he notes, is dominated by simulation and rules of thumb. But rules of thumb are far from exact, and simulations based on sets of equations become increasingly difficult to produce as wireless implementations get more complex. (Consider, for instance, a wireless networking provider that tries to manage connections, client device power usage and interference reduction in a metropolitan area with hundreds of thousands of users, some of whom are standing still while others are rapidly moving down the highway.) MacKenzie says he believes that game-theory-based models of such environments may one day provide more precision than rules of thumb, while not requiring the complexity of detailed simulations. The ultimate goal is to create analysis tools that wireless engineers could use when designing and managing their networks.

Wireless providers have yet to come knocking on MacKenzie’s door. (“We’re not quite ready to answer the types of questions they’d like to ask,” he says.) But the National Science Foundation was interested enough to give the professor a $400,000 grant to continue his research.

MacKenzie can’t say for certain when his theories will yield practical results, although he says that tools are at least a year or two away.