Application Of Workload Balancing Algorithm In Mobile NetWorks Optimization

With the explosive growth of mobile traffic demand,high deployment expense,and scarce frequency resource,efficient network planning algorithm becomes more and more critical.To satisfy the constantly increasing demand for communication capacity,one of the most important methods is densely deploying cells with different transmission powers to improve area spectrum efficiency,and it will inevitably increase the number of base stations,as well as the cost of the infrastructure.In order to handle such situation,C-RAN(Cloud Radio Access Network)has been proposed based on centralized cloud computing techniques.The core idea of the C-RAN is the aggregation of base band units into the cloud,which greatly decreases the capital expenditure and operating expenditure of the system.On the other hand,the idea of workload balancing has been applied for the purposes of increasing the operation efficiency of the system and cutting down on costs more and more frequently.If we apply workload bal-ancing technique to the network planning of heterogeneous networks or C-RAN,we can enhance the utilization rate of system's capacity without extra cost on construction or operation.In conclusion,the idea of workload balancing will be a cheap and efficient way of improving the system's usage rate and decreasing cost,no matter which network architecture will be adopted.The contributions of this thesis have been summarized as below:1.We introduce workload balancing algorithm to the planning of heterogeneous networks.Through balancing the traffic demand faced by each cell,the utilization rate of the system's capacity has been enhanced.The key to the proposal can be formulated as a convex optimization problem.Simulation results prove that workload balancing algorithm can largely increase the system's capacity available and improve the QoS(Quality of Service)of users.2.We propose a C-RAN deployment scheme based on workload balancing.By balancing the traffic demand faced by each RRH,not only is the utilization rate of system's capacity increased,but the high demand on the bandwidth of backhaul is also alleviated,decreasing system's capital expenditure.The core idea of the proposal can be formulated as a convex optimization problem.Numerical results indicate that workload balancing algorithm can increase the capacity usage rate of the system,and it is quite robust.To sum up,the idea of applying workload balancing algorithm to the network planning of heterogeneous networks and C-RAN proposed in this thesis can efficiently enhance the utilization rate of system capacity,together with decreasing the cost of the whole system,and improving the communication quality of users.It throws insights into a completely new angle of networks planning.