For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Performance Evaluation of the Centralized Spectrum Access Strategy with Multiple Input Streams in Cognitive Radio Networks
Yuan ZHAO Shunfu JIN Wuyi YUE
IEICE TRANSACTIONS on Communications
Publication Date: 2014/02/01
Online ISSN: 1745-1345
Print ISSN: 0916-8516
Type of Manuscript: Special Section PAPER (Special Section on Technologies for Effective Utilization of Spectrum White Space)
cognitive radio networks, spectrum access, multiple channels, Markov chain, optimization,
Full Text: PDF>>
In this paper, we focus on a centralized spectrum access strategy in a cognitive radio network, in which a single licensed spectrum with one primary user (PU) and several secondary users (SUs) (multiple input streams) are considered. We assume the spectrum can be divided into multiple channels and the packets from variable SUs can arrive at the system simultaneously. Taking into account the priority of the PU, we suppose that one PU packet can occupy the whole licensed spectrum, while one SU packet will occupy only one of the channels split from the licensed spectrum when that channel is not used. Moreover, in order to reduce the blocking ratio of the SUs, a buffer with finite capacity for the SUs is set. Regarding the packet arrivals from different SUs as multiple input streams, we build a two-dimensional Markov chain model based on the phase of the licensed spectrum and the number of SU packets in the buffer. Then we give the transition probability matrix for the Markov chain. Additionally, we analyze the system model in steady state and derive some important performance measures for the SUs, such as the average queue length in the buffer, the throughput and the blocking ratio. With the trade-off between different performance measures, we construct a net benefit function. At last, we provide numerical results to show the change trends of the performance measures with respect to the capacity of the SU buffer under different network conditions, and optimize the capacity of the SU buffer accordingly.