Research On Resource Allocation In Femtocell Networks Based On Interference Management

Femtocell technology has the advantages of low cost and low consumption, which can improve the indoor covering performance and system capacity. It has been considered as the important development direction for future wireless communication network. However, the deployment of femtocell in macro cellular network makes the interference scenarios more complex, and increases the difficulty of the whole system's interference management and resource allocation. Therefore, the research on interference management and resource allocation for femtocell is very important.Aiming at the problem that present clustering algorithms have a wide range to search the optimal clustering scheme because of no femto base stations grouping before clustering, which leads to much higher computational complexity, this thesis proposes a clustering algorithm reducing the computational complexity by grouping femto base stations according to interference graph and adjacent matrix so as to narrow the optimal scheme searching range. Aiming at the problem that the existing power control schemes improve the energy efficiency at the expense of decreasing the quality of service and are only suitable for single service, this thesis proposes an energy-efficient resource allocation scheme regarding weighting energy efficiency which takes user interruption problem, different service requirement and network spectrum efficiency into consideration as individual fitness function. According to proposed above clustering algorithm and power control scheme, this thesis designs a high energy-efficient resource allocation scheme based on clustering. The scheme firstly clusters femto base stations to cancel the co-channel interference among femtocells and improve network spectrum efficiency; then on the basis of clustering results, the frequency division and sub-channel allocation of femtocells are performed; finally, the power control of femtocells is performed to increasing energy efficiency. Simulation results show that,compared to existing clustering algorithm with no femto base stations grouping before clustering, the proposed clustering algorithm can gain comparative user average data rate while having lower computational complexity; compared to the resource allocation scheme simply based on energy efficiency, the proposed high energy-efficient resource allocation scheme based on clustering has higher average data rate while having comparative energy efficiency.The static independent channel configuration simply allocates the frequency resource into two parts, which lowers the frequency efficiency. Therefore, this thesis proposes a dynamic and static combination frequency partition scheme based on feedback of users' satisfaction. The proposed scheme adopts independent channel configuration to initialize frequency division according to the ratio between the number of macro base station and femto base station, and then the macro-base station and femto base station performs user scheduling and resource allocation based on clustering within available frequency, respectively. Finally, calculating the usage status of frequency resource and user transmission rate, and then based on the frequency surplus and user's satisfaction, adjusting the number of frequency resource of macro users and femto users by utilizing coarse adjustment and fine adjustment. Simulation results show that, the proposed scheme can improve system's frequency efficiency, users' satisfaction and average throughput compared with the static independent channel configuration.