انجمن کامپیوتر ایران

In traditional cooperative spectrum sensing such as OR-rule or AND-rule, secondary user (SU) must maintain coordination based on a fusion center. In this paper, we propose the weighted average consen- sus, based on Learning Automata(LA), for fully distributed cooperative spectrum sensing without fusion center. At the first stage of the pro- posed scheme, each SU makes measurement about presence of primary user (PU) at the beginning of each time slot, then communicates with local neighbors to exchange information to make the final decision and update its weight using a LA based algorithm. Simulation results show that the proposed scheme has better performance than the non weighted consensus and existing weighted consensus scheme. Also, the convergence time of the proposed scheme is less than the existing weighted consensus and almost equal to non weighted consensus scheme