Application Study On Game Theory For Cooperative Power Allocation

In order to satisfy people’s improving quality of mobile communication requirements, MIMOtechnology is proposed;however, due to the limitations in the practical application of MIMOtechnology, cooperative communication technology is widely used in power allocation of newgeneration mobile communication systems.There are many ways about power allocation amongcooperative communications, and most applications are centralized allocation strategies. However,centralized allocation strategy needs a large amount of channel state information,and requires alarge amount of information exchange state between the source node and the relay node, thedestination node, which leads to a huge signaling overhead of the system. Taking into account thecharacteristics of distribution in cooperative communication,the game theory is applied to thepower distribution cooperative communication.Details are as follows:(1)Firstly, this paper analyses and comparises the basic cooperative communication relayselection and power allocation algorithm，and research application of game theory in the field ofcooperative communications.(2)This paper explains the process of establishing of Stackelberg model laborately,and based onthe classical Stackelberg model,improves the game.Then this paper analyses its new game model,and finally prove the existence of a balanced game.(3)The paper proposes improved algorithms for the shortcomings of traditional StackelbergModel.Relay selection threshold of the traditional Stackelberg model is relatively low, with theincreasing number of cooperative relays, many will be selected, and the complexity of thealgorithm itself is relatively high, which will bring the system heavy signaling overhead.Twoselection algorithms are proposed for limiting the number of cooperative relay, and this paperanalyzes the influence of relay selection threshold on the number of cooperative relays.(4)This paper transforms the Stackelberg model into the form of centralized allocationstrategy,and then compares it with the traditional centralized allocation strategy.Then this paperestalish a Nash game model, and the results are compared with Stackelberg model.AlthoughStackelberg model corresponding to the Nash game model has higher complexity, it does not needfull CSI, and can reach the data transmission rate of centralized allocation strategy.