Power Control In Cognitive Radio Networks: A Game-Theoretic Perspective

The 21st century is an economic globalization and global information age. The development of wireless communication techniques and higher data transmission re-quired by the emerging wireless services make wireless spectrum resource scarce. Actually, spectrum scarcity is mainly due to current static spectrum allocation pol-icy, not caused by the available spectrum itself. Under the framework of fixed radio spectrum management, most of the allocated spectrum is in the low utilization state. How to efficiently use resource with the aid of cognitive radio is promisingly becoming one of the effective techniques to solve the lack of resources and spectrum scarcity. Recently, spectrum sharing has been paid a significant attention from the cognitive radio community; meanwhile, it is widely applied in practice. However, the new tech-nical solution requires a new design and analysis techniques, which will help design a distributed and flexible resource management scheme. Based on game theory, we investigate distributed resource allocation and dynamic spectrum sharing in cognitive wireless networks, and contributions are summarized as follows:1. We investigate the uplink power control in cognitive radio networks. Nash equi-librium is employed as the optimal design guideline for the power control. A pricing function based on signal-to-interference-plus-noise (SINR) is proposed, meanwhile, with the consideration of fairness among cognitive users, we finally achieve a novel utility function. Hence, a non-cooperative power control game model is formulated on the base of this utility function. Mathematically, we con-clude the existence and uniqueness of the Nash equilibrium solution for this proposed game model. Moreover, we present a joint rate and power control scheme in the case of requirements of diverse needs of transmission rate of different cognitive users. Simulation results show the performance Pareto im-provement of the proposed pricing schemes.2. A joint strategy design of dynamic spectrum leasing (DSL) is studied. We pro-pose a pricing-based dynamic spectrum leasing (PBDSL) model, as well as we fully consider both economical and technical perspectives of conventional dynamic spectrum sharing. Meanwhile, both the spectrum revenue of the pri-mary users (PUs) and quality of service guarantee of secondary users (SUs) are considered in the hierarchical Stackelberg PBDSL game model. Primary users as players enter this game, distributed spectrum pricing schemes and optimal transmission strategies are proposed for the PUs and SUs, respectively. Numer-ical results demonstrate the convergence property of the interactive decision-making process, and verify average throughput, energy efficiency and spectrum efficiency improvements, compared with the Nash equilibrium solution (NES)-based approach.3. We investigate the hierarchical dynamic spectrum sharing scheme with asym-metric context and strategy information structure among multiple secondary users. A Stackelberg capacity maximization game (SCMG) is formulated and used to capture the multiple channels sharing issue of multiple secondary users. We study the optimal decision making when there is information asymmetry phenomenon among different secondary users. More importantly, contrary to previous works, a practical multi-leader and multi-follower coexisting with each other to share multi-channel scenario is studied, and we prove the equiva-lence of SCMG model to the variation inequality model, further, the existence and uniqueness of equilibrium solutions are derived. Simulation results show the guarantee of convergence of the proposed distributed algorithm, which can achieve optimal solution with a low implementation complexity.4. We employ cooperative game to achieve the best tradeoff of network efficiency and user fairness. The multiple constraint conditions coupled power control problem is formulated as a cooperative Nash bargaining power control game (NBPCG) model. A SINR-based utility function is designed for this game model, which not only reflects the spectrum efficiency of the CRN, but also complies with all the axioms in the Nash theorem and hence facilitates efficient algorith-mic development. The existence, uniqueness and fairness of this game solution are proved analytically. To deal with the I PCs where the power control decisions of all users are coupled, these I PCs are properly transformed into a pricing func-tion in the objective utility. Theoretical analysis and simulations are provided to testify the effectiveness of the proposed cooperative game algorithms for effi-cient and fair power control in CRNs.