Performance Enhancement of Dynamic Spectrum Access via Channel Reservation for Cognitive Radio Networks

In cognitive radio networks models, quality of service (QoS) of primary users (PUs) must be assured. Dynamic spectrum access is a paradigm by which a radio system adjusts dynamically the use of convenient spectrum holes. In this paper, a secondary user reserved channel (SU-RC) model is proposed. SU-RC model introduces the use of a new SU reserved channel infrastructure to enhance QoS of SUs. Furthermore, SU-RC improves the efficiency of network by reducing the blocking probability and the forced termination probability of SUs. The proposed algorithm is significantly adaptable by deducing the optimal number of reservation channels. A reasonable balance between the success probability of channel selection and average number of channel switching is accomplished. Furthermore, this algorithm demonstrates the impact of PU’s interference either behind or inside influenced region on SUs. Simulation results show that, by applying the SU-RC algorithm with a preferable number of reservation channels, the number of channel switching is still very close to that of the network without external SU-reserved channel. For example, for the case if SU is inside the PU’s influenced region, when λp=20\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda_{p } = 20$$\end{document}, Pf=0.05\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{f} = 0.05$$\end{document} has constant the optimal number of reservation channels nopt=2\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$n_{opt } = 2$$\end{document} for both cases either with or without the existence of reserved channel. Furthermore, since λp=10\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda_{p } = 10$$\end{document}, Pm=0.05\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$P_{m} = 0.05$$\end{document}, the average number of channel switching S¯t\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{ S} \left( t \right)$$\end{document} is equal to 1.005 in case of without existence of the external reserved channel whereas S¯t\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{ S} \left( t \right)$$\end{document} is approximate 1.0275 which is regarded an in considerable increase of S¯t\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\overline{ S} \left( t \right)$$\end{document} is about 0.0225.