Power control algorithm based on new-style utility function in cognitive system

A utility function with adjustable parameters is proposed and just only related to SINR threshold, and it is quite suited to power control analysis of cognitive system as a result of avoiding conventional utility function design related to specific modulation. A non-cooperative game power control algorithm is brought forward in the form of utility function. The proposed game model is in accordance with next top model game by proving, and guarantees the existence of Nash equilibrium power solution in theory, but it will cause that the Nash equilibrium solution isn´t satisfied with Pareto optimality due to the selfish and rational behavior of cognitive users, so there are still opportunities to improve on Pareto. Therefore, the validity of Nash equilibrium solution is further improved combining with the definition of cost function. The simulation results show that the proposed algorithm comparing with the reference algorithm can obtain the same system utility with little transmission power. Meanwhile, the algorithm based on cost function indeed effectively improves the validity of original model Nash equilibrium power solution.