Dynamic Spectrum Access in Cognitive Radio Network using Static Neighbor Graph based prediction

Abstract

In Cognitive Radio Network (CRN), secondary users are allowed to access the licensed spectrum bands when it was not used by the primary user. The activities of the secondary users such as “how long to sense” and “how long to transmit” make a considerable intrusion into the quality of service of the secondary user due to dynamic spectrum access. Dynamic spectrum access techniques improve the efficiency of spectrum utilization. To alleviate the processing delay involved in spectrum utilization, spectrum prediction technique has been proposed for CRN. In this paper, an approach called Static Neighbor Graph (SNG) based prediction used for the prediction of the occupancy status of a primary user. Besides the prediction, the secondary user evades the sensing duty on busy channels which makes a considerable reduction in sensing time and energy consumption significantly. Along with SNG, the Carrier Sense Multiple Access/Collision Avoidance (CSMA/CA) is suggested for the coordination of the channel access to the secondary network. The performance of this proposed work has been illustrated and validated in terms of throughput, packet delivery ratio, and channel utilization.