Özet:
The underutilization of spectrum coupled with developments in network technologies has prompted a number of proposals for managing spectrum. Dynamic spectrum access radio technology, which is based on cognitive radio technology, promises to increase spectrum sharing and thus overcome the lack of available spectrum for new communication services. It allows unlicensed secondary systems to share the spectrum with the licensed systems. In this dissertation, different architectures are investigated in a cognitive radio environment. The considered network is assumed to consist of multiple primary service providers which have some unutilized bandwidth, and multiple secondary users that require bandwidth. Secondary users are assumed to pay the primary service providers for short term usage of their available spectrum bands; which is referred as the spectrum trading. The proposed architectures all aim at establishing a framework where each type of users satisfies with the services. As each new entrant secondary user creates interference on the incumbent users, controlling the power emission in a cognitive radio network is crucial in spectrum trading. Furthermore, proposed architectures examine the unit spectrum prices that primary service providers set in the multiple-seller and multiplebuyer environment. Modeling the competitive relationships among network elements as games ensures analyzing all elements’ behaviors and actions in a formalized way. The existence of various network elements that want to maximize its own profits makes the problem very complex, with usually conflicting objective functions. Therefore, the proposed pricing models have their basis on the game theory in order to deal with the severe competition in spectrum trading markets.
