A Virtual Grid Sensing Strategy Based on Three Regions of Cognitive Radio System

In this paper, a virtual grid sensing strategy based on three regions of cognitive radio system is investigated. The coverage of a primary user could be divided into three regions and every secondary user could work in underlay or overlay mode according to the region it locates in. In traditional sensing strategy, the system should guide every secondary user to sense all channels over sensing duration and make decision of which region one secondary user locates in simply based on the secondary user's sample data. In proposed virtual grid sensing strategy, a virtual grid network and virtual secondary users are introduced. The system would make full use of the connections between one virtual secondary user and its neighbors and classifies them into different states. Based on the classification, the system could find the secondary users close to the region boundaries more accurately. The system would only track the secondary users near the boundaries in the dynamic environment while release other secondary users' sensing times for transmission. Using this strategy, the system could keep in a high throughput while have a low probability of misdetection and false alarm. Simulation results show that our proposed strategy has a better performance than the traditional sensing strategy.