Research On User Offloading Strategy In Heterogeneous Cellular Networks

Traditional cellular networks can't meet the rapid growth of the mobile data service needs because of the limited transmission capacity,so cellular network operators suffer tremendous pressure in the face of the explosive growth of data traffic.Heterogeneous networks have been widely deployed as a cost-effective way to enhance the network capacity.However,due to the huge power difference between the macro base stations and the low power nodes,the cell selection strategy based on the traditional homogeneous cellular network causes the load distribution unbalanced in heterogeneous networks,resulting in congestion within macrocells and resources underutilized within low-power cells,which greatly limits the performance of heterogeneous networks.Offloading some users from a congested macrocell to a resource-rich low-power cell is an effective way to improve the performance economically.User offloading strategies need to make appropriate user offloading decisions according to the actual state of the network such as user distribution,load situation,base station operation mode and so on,and also need to consider the multiple factors related to the network performance such as interference coordination,resource allocation,commercial cost and profit.Therefore,the focus of this thesis is to research the user offloading strategy considering the multiple factors that affect the network performance in different scenarios.Aiming at the problem of load unbalance in the same operator scenario,we study the user offloading and interference coordination strategy.Firstly,a joint utility optimization problem considering the parameters of user association,almost blank subframes(ABS)ratio and resource allocation ratio is formulated.Then we use Gauss-Seidel block coordinate decomposition method to decompose the joint problem into three sub-problems and solve them in an iterative manner.Furthermore,we propose a heuristic algorithm with low complexity based on the intermediate analytical results to solve the user association problem.The simulation results show that the proposed scheme can greatly improve the users' edge throughput and fairness compared with the cell range expansion(CRE)scheme.Aiming at the problem of load unbalance in the different operator scenario,we study the user offloading and cost coordination strategy focusing on two contradictory needs that the cellular service provider wants lower leasing cost and larger amount of offloading traffic.Firstly,a reverse auction problem aiming to get the minimum difference between offloading costs and gains is formulated.Then we propose reverse auction based optimal offloading mechanism and spectrum efficient greedy offloading mechanism with low complexity,and they both satisfy incentive compatibility and individual rationality.The simulation results show that the optimal offloading mechanism can balance the offloading cost and the offloading gain well,and the greedy offloading mechanism can ensure the compromise between offloading cost and offloading gain with low complexity.