Recent advances in software-defined radios can potentially enable wireless applications to locate and exploit underutilized radio spectrum. This has motivated a re-evaluation of policies for spectrum allocation and management. Changes in policy will encourage the use of new and more efficient mechanisms for dynamic spectrum sharing. This project is studying the design and performance of such mechanisms. Two policy scenarios for spectrum management are being considered. In the exclusive use model, a piece of spectrum is managed exclusively by the owner or licensee, who can lease the spectrum to others. In the commons model the spectrum can be used by anyone provided that they comply with an accepted protocol or standard (as in unlicensed bands). This project is developing mechanisms to facilitate spectrum sharing for each of these scenarios, and attempting to characterize how the performance (e.g., total utility) achieved by various mechanisms depends on constraints on information exchange, computation, and incentives. To accomplish this, modeling and performance evaluation techniques for wireless communications and networking are being combined with economic-based techniques, such as game theory and mechanism design. It is expected that this work will provide basic insight into the behavior of different spectrum sharing mechanisms along with the trade-offs associated with different approaches.
