A Novel Spectrum-Scheduling Scheme for Multichannel Cognitive Radio Network and Performance Analysis

In this paper, we develop an opportunistic spectrum-scheduling scheme for multichannel cognitive radio networks. In the proposed scheme, the primary user activity and channel quality (signal-to-noise ratio (SNR) at the receiving secondary user) vary on a slot-by-slot basis. The scheduling is performed at the beginning of the frame, which consists of multiple slots. The scheduling algorithm estimates the expected number of packets that can be transmitted over the frame by each secondary user for each licensed channel. We present a Markov chain formulation to calculate the expected number of packets that can be transmitted over the frame for a secondary user that corresponds to each licensed channel. Based on these expected packet transmissions, a central scheduler allocates the licensed channels to the secondary users. The objective of the scheduling algorithm is to allocate the licensed channels to maximize the aggregate throughput of the secondary users. Compared with the existing dynamic spectrum access schemes, the proposed scheduling scheme incurs smaller scheduling overhead and achieves higher throughput. We also present an analytical framework that models the proposed algorithm using the Markov chain analysis. The performance of the proposed scheduling scheme is evaluated in terms of the average user throughput and blocking probability.