A New Framework for Spectrum Sharing based on Multiagent Systems and Petri Nets

International audienceRecently deployed wireless infrastructures follow static spectrum assignment policies, in which the spectrum is assigned to the licensed (or primary) users for long time durations. These static assignments create huge empty spectral resources which can be utilized dynamically. This dynamic access is not so obvious to perform and many related works have been done with their pros and cons. Therefore, this paper presents a cooperative framework trying to solve the consof the existing approaches for dynamic spectrum access in cognitive radio networks, where the cognitive radio (or unlicensed) devices may use the currently unutilized spectrum dynamically and opportunistically. The proposed solution works on the principle of a multiagent system that allows the cooperation amongst several participating devices. The key aspect of our design is the deployment of agents on each of the primary and cognitive radio device that cooperates in order to have a better and dynamic use of the spectrum. For cooperation, contract net protocol is considered allowing spectrum to be dynamically allocated by having a series of messages exchange. Due to the concurrent, distributed and autonomous nature of our approach, Petri nets are adopted to model the behavior of cooperative framework. Based on Petri nets, we study the internal message passing and spectrum allocation mechanism between the participating primary and cognitive radio users and identify the information needed by the agents to make cooperative decisions. Then, using empirical results, we show that our approach achieves up to 80% of the whole utility within the span of few messages, and provides an effective mechanism for distributed spectrum allocation