Game Theory Based Network Selection And Dynamic Resource Allocation Strategy

With the development of wireless communication, high speed communication becomes more common. As the3GPP enables full frequency reuse, inter-cell interference becomes the primary cause of low transmission speed. Coordinate Multipoint (CoMP) can reduce or eliminate inter-cell interference by introducing cooperation among BSs (Base Station). Field trail have proved great improvement on cell edge users using CoMP. However, it is impractical to cooperate among all BSs, how to cluster BSs raises as an main question to researchers. Furthermore, traditional power allocation algorithm generally require a central unit to process calculation, while CoMP is mainly distributed. It is difficult to apply this algorithm directly in CoMP. Thus, some distribute power allocation algorithm should be discussed. The main work of this dissertation is included as following.First, a novel dynamic sub-carrier allocation algorithm is proposed. In this algorithm, we design a new gain measure coefficient to select sub-carrier and restrict cluster size dynamically, which improves frequency efficiency and energy efficiency. The main process of algorithm is divided into four phases; firstly users should estimate its own user type based on channel condition. Secondary, BS should gather all cell edge users’candidate cell and calculate the best sub-carrier and cluster scheme in cell. Thirdly BSs iteratively negotiate with others to select the best scheme among system. Fourthly BS informs cooperating BSs and edge user of chosen scheme. Simulation results confirm that proposed scheme outperforms the existing scheme in terms of frequency efficiency and network selection’s fairness. Second, a green distribute power allocation based on game theory is proposed to improve energy efficiency and frequency efficiency. The main process is divided into two phases:first each BS should decide the maximum transmission power based on equal power allocation or water filling algorithm. Secondly BS iteratively calculate new transmission power and share with others till no more change, which means Nash Equation reached. Simulation results show that, the proposed algorithm performs well in energy efficiency and frequency efficiency.