Resource Allocation For Users In Cognitive Femtocell Networks

Femtocell base stations deployed by users greatly improve the communication quality of indoor users,while result in the problem of co-channel interference at the same time.Due to the low algorithm complexity and well-adapted behavior in self-organizing heterogeneous networks,coalition formation game and matching strategy are widely used in femtocell networks.This paper mainly studies the effects of these two algorithms on subchannel allocation.The main contents and innovations are summarized as follows:Aiming at the diversity of alliance strategies of coalition formation game in subchannel allocation schemes,this thesis studies the effect of alliance strategies,including player identity,the number of coalitions that players can try to join,the order in which players join coalitions,and the conditions for players joining coalitions,etc.,on the execution results,including throughput,computational complexity and communication overhead,of different coalition formation games.Simulation results testify the effectiveness of coalition formation game in reducing the co-channel interference.When the density of femtocell base stations in the network is 1/100m2,the cooperative strategy in the coalition formation game can improve the total throughput of femtocell users by about 37.54%.In addition,the simulation results show that,NeighborMax algorithm introduces the idea of neighbor coalition proposed by existing literatures performs best overall.Because threshold factor determines the implementation effect of NeighborMax algorithm,the implementation results of NeighborMax algorithm under different interference threshold factors are simulated and analyzed in this thesis.And the optimum disturbance threshold is obtained.Besides,since the hybrid spectrum sharing mode can make more efficient use of channel resources,this thises studies the impact of the number of subchannels in overlay shareing mode on the throughput of femtocells networks,when adopting NeighborMax algorithm.Simulation results show that,when the number of sbuchannels in overlay sharing mode increases more than 4,the network throughput will increase more obviously in the simulation environment.Based on the coalition formation game,in order to further improve the channel utilization of the femtocell network,this thesis proposes a fair many-to-many matching algorithm,which is suitable for subchannel allocation within a coalition,to allocate multiple subchannels for each femtocell user in each coalition.Simulation results show that,compared with the many-to-one matching game,many-to-many matching game proposed in this thesis can significantly improve the channel utilization rate of the network and the total throughput of femtocell users,thongh increasing a certain amount of computational complexity and network over-head.In particular,when the number of femtocell base stations in the network is 40 and the number of available subchannels is 41,compared with adopting the many-to-one matching game,the throughput of femtocell base stations can be improved by about 346.74%by adopting the many-to-many matching game.