Study On Resources Optimization And Interference Control In Two-tier Femtocell Networks Based On Game Theory

Femtocell also called home base station, which is low-cost low-power short-rangewireless data access points installed by terminal users in the building or at home. It canimprove the indoor coverage and get better data and voice service for the cellularcommunication. However, it causes serious interference problem when the femtocells aredeployed. Furthermore, there are some factors also affect the network performance, suchas the spectrum allocation, the access priority, the installation of the intensity and differentQos requirement. In the spectrum-sharing two-tier femtocell network, it is an effectivemathematical tool by using game theory to solve the problem of optimization andinterference control.Firstly, considering the problem that the centralized power control method andjointing control power and optimize the femtocell network performance by power controlalgorithm are difficult to implement in the previous literatures, the distributed powercontrol based on non-cooperative game in two-tier femtocell network is formulated in thispaper. The aggregate interference threshold, which the macrocell can be tolerant, isregarded as the allocated resource. The femtocell users compete with each other andpurchase it by the way of non-cooperative game. Then we analyze the existence anduniqueness of Nash equilibrium in the non-cooperative game, and the distributed poweradjustment algorithm and user selection algorithm also be given. The simulation resultsprove the validity of algorithm, and the system throughput can be improved effectively.Secondly, in view of the cross-tier interference and co-tier interference need to besolved when the femtocells are deployed densely, the distributed power control based onStackelberg game in the two-tier femtocell network be studied. In the Stackelberg game,the macrocell base station is the leader and the femtocell base station is the follower. Toensure the Qos of macrocell, a punishment function which is the utility function of themacrocell base station is proposed, and we can get optimal interference threshold byoptimize the macrocell utility. Under the condition of the interference threshold, afictitious game is given to control the co-tier interference among femtocells. Furthermore, a modified distributed power control algorithm which can connect the whole game processis proposed. The simulation results show that the method is better than the method of non-cooperative power control, and the whole network performance can be improved.Finally, assume that the powers of femtocells are constant, the co-tier interferenceamong femtocells can be reduced by sub-channel allocation. Thus, the bandwidthallocation with NBS(nash bargaining solution) in cluster-based femtocell networks isstudied. The interference weight graph is constructed by the interference relationshipamong femtocells firstly, then the femtocell clusters are established. To optimize thenetwork performance, the minimal requirement of femtocell need to be satisfied, and theresource allocation based on NBS allocate different bandwidth to each cluster isformulated. Simulation results show that the NBS based resource allocation which cansatisfy the demand of femtocells is better than average resource allocation method, and theSINR(signal to interference plus noise ratio) of the femtocell network can be improveefficiently.