Spectrum Sharing Strategies For Cognitive Radio Networks

Cognitive radio (CR) is a breakthrough in the wireless communication field. As the most promising paradigm to achieve efficient use of the spec-trum resource, cognitive radio technologies allow secondary users (SUs) to share licensed spectrum with primary users (PUs) by sensing the sur-rounding wireless environment, adjusting their parameters dynamically and therefore avoiding incur harmful interference to the PUs. Since PUs and SUs are rational but selfish users, to further improve the spectrum usage efficiency, we analyze the spectrum sharing problem in cognitive radio net-works with game theory models and the major contributions are outlined as follows.1. By assuming a PU is willing to lease its bandwidth for a fraction of time to SUs in exchange for selective cooperation, we propose a Stackel-berg game model to describe the dynamic interaction between the PU and SUs. In our model, the PU plays a game with each potential SU and selects the best one to cooperate with according to a specific criteria. We derive explicitly the condition under which an SU would like to join the game, and prove that the game always possesses a unique Nash equilibrium (NE). Moreover, we propose a simple distributed algorithm for practical imple-mentation of the model. Simulation results verify that our approach can benefit both the PU and SU in terms of utility and transmission rate, which demonstrates that our proposed model is a practically appealing candidate for CR networks.2. In cognitive radio networks, the primary user (PU) can lease a frac-tion of its licensed spectrum to the secondary users (SUs) in exchange for their cooperative transmission if it has a minimum transmission rate re-quirement and is experiencing a bad channel condition. When SUs belong to the same organization and can work as a group, how to stimulate them to cooperate with the PU and and how to coordinate the usage of rewarded spectrum among these SUs after cooperation are critical challenges. In this paper, we propose a two-level bargain framework to address the aforemen-tioned problems. In the proposed framework, the interactions between the PU and the SUs are modeled as the upper level bargaining game while the lower level bargaining game is used to formulate the SUs’decision mak-ing process on spectrum sharing. We analyze the optimal actions of the users and derive the theoretic results for the one-PU one-SU scenario. To find the solutions for the one-PU multi-SU scenario, we put forward a re-vised numerical searching algorithm and prove its convergence. Finally, we demonstrate the effectiveness and efficiency of the proposed scheme through simulations.3. How to stimulate the SUs to play cooperatively with the PU is an important issue. In this paper, we propose a reputation-based indirect reci-procity game, where SUs choose how to help PUs relay information and gain reputations, based on which they can access a certain amount of va-cant licensed channels in the future. By formulating the decision making of an SU as a Markov decision process (MDP), we obtain the optimal action rule, according to which the SU will use maximal power to help PU relay data if the channel is not in an outage, and thus greatly improve the PU’s QoS as well as the spectrum utilization efficiency. Moreover, we prove the uniqueness of stationary reputation distribution and theoretically derive the condition under which the optimal action rule is evolutionarily stable. Finally, simulation results are shown to verify the effectiveness of the pro-posed scheme.4. In cognitive radio networks, one of the critical issues for fully uti-lizing the spectrum resources is how the SUs obtain accurate information about the PUs’activities and make right decisions on which channel to access so as to avoid competition from other SUs. In this paper, we consid-er a scenario where SUs sense channels simultaneously and make access decisions sequentially. Considering negative network externality, we for-mulate SUs’learning and decision making process as a Chinese restaurant game. In the proposed game, SUs build the knowledge of the PUs’activi- ties by their own sensing and the information learn from other SUs. They also predict the subsequent SUs’ decisions to maximize their own utilities. We analyze the interactions among SUs and study specifically the impact of SUs’initial belief, sensing accuracy and channel quality on their deci-sions. We also derive the theoretic results for the two-user two-channel case and extend the results to the multi-user multi-channel case. Finally, we demonstrate the effectiveness and efficiency of the proposed scheme through simulations.