Research On Power Control Algorithm Based On Non-cooperative Game

With the rapid development of wireless communication technology, the demand for wireless communication is increasing and the available spectrum resources are getting less and less, meanwhile, the utilization of the current spectrum allocation is extremely inefficiency. Cognitive radio technology adopts dynamic spectrum allocation method, which can realize frequency spectrum sharing and improve spectrum utilization efficiently. The transmission power control for the cognitive users is the key prerequisite technique to share the spectrum. This paper mainly investigates the power control for the multi-user cognitive network, and further optimizes the power control under a specific QoS requirement by using game theory. The main work is as follows:At first, for the obvious defect of NPG algorithm that it working without considering the inter-user influences and NPGP algorithm can not guarantee the fairness of cognitive users, this paper studies NPG-IT power control algorithm based on interference threshold and validates all three algorithms via the simulations.Simulation results shows that comparing with NPG and NPGP algorithm, NPG-IT algorithm not only reduces the transmission power of users, but also improves the users’ utility value, and reduces the lowest bound of SINR to guarantee normal communication of cognitive network, improves the fairness of cognitive users.After that, the paper investigates NPG-NL power control algorithm based on network load, this algorithm can adjust cost function of cognitive users according to the real-time network load. Specifically, the transmission power, user utility, SINR and efficiency-power-ratio are adequately analyzed, and the NPG-NL is validated under different user number via simulations. The simulations demonstrate that, the convergence rate of NPG-NL is quickly enough which is able to satisfy the real-time requirement of power control algorithm. In addition to that, the transmission power can be saved, and the user fairness is also enhanced significantly.