Power Control Algorithm Based On Non-cooperative Game In Cognitive Radio Networks

With the rapid development of wireless communication technology and the increasing need of wireless communication, the shortage of radio spectrum resources becomes more and more serious. Cognitive radio technology is considered to be an effective scheme to solve this problem. Many researchers and institutions around the world have comprehensively and deeply studied this technology. Many new research achievements have been obtained in the areas of spectrum sensing, spectrum allocation and power control of cognitive radio. The research of power control algorithms based on non-cooperative games in cognitive radio is of important guiding significance for resolving the problem of power control in cognitive radio.The research scheme of combining theoretical analysis and numerical simulation is adopted to improve the power control in cognitive radio. The existence of Nash Equilibrium points is proved for the models of non-cooperative power control game and pricing-function-based non-cooperative power control game model. The effectiveness of these two game models are evaluated through simulations, and the properties in transmitting power and signal-to-interference ratio of the simulation results are compared. Simulation results show that, both NPG and NPGP models can be used to control transmitting power. Compared with NPG model, NPGP model can make each user work with a relatively lower transmitting power and obtain higher utilities. However, fairness of users'signal-to-interference ratio in NPGP model is worse than that in NPG model.Based on the original power control game algorithm, a new cost function is proposed for the purpose of application to enhance the convergence rate of the game algorithm and reduce the transmitting power of all the cognitive users. The problem is solved recursively, and the convergence of the recursive algorithm is proved. Under different scenarios regarding the distribution of users, the properties of the proposed algorithm (such as convergence rate, transmitting power and signal-to-interference ratio) are evaluated through simulations, and compared with those of the SIR balancing algorithm and Koskie-Gajic algorithm. Simulation results show that, whether in the cases of fixed users, or in the cases of randomly-distributed users, the properties in transmitting power and signal-to-interference ratio of the proposed game algorithm are better than those of the original game algorithm. The proposed game algorithm has an obvious effect in reducing power consumption and saving energy.