Resource Allocation For Heterogeneous Networks

As a novel candidate technology in the 5th generation wireless networks,heterogeneous network is proposed to increase network throughput and coverage,and reduce energy consumption.In the homogeneous networks,the transmission power and coverage of each base station(BS)is similar.Nevertheless,macro cell and small cell BSs with different transmission power and processing capability are deployed in heterogeneous networks to meet various communication demands.The small cell BSs include picocell BSs,femtocell BSs and relays.Picocell BSs and femtocell BSs are connected to the network by wired backhaul and relays are connected to the network by wireless backhaul.A heterogeneous architecture brings in a rich topology,but the deployment of different low power BSs over existing macro BSs coverage causes severe interference,which poses new challenges on interference management and resource allocation.Therefore,it is imperative to develop new resource allocation algorithms in heterogeneous networks.The main contributions of this thesis are summarized as follows:Firstly,for the multi-cell Orthogonal Frequency Division Multiple Access(OFDMA)heterogeneous networks,we develop a novel system model considering both multi-association and co-channel deployment.We focus on two problems for this model.The first one is joint optimization of user association and subchannel allocation for a fixed power allocation scheme.We transform it into an equivalent bipartite graph matching problem and obtain the globally optimal solution by the Hungarian algorithm.The second problem is power allocation for fixed user association and subchannel allocation.We transform it into a series of convex problems based on the difference of two concave functions approximation(DCA)method.Then we propose a low complexity algorithm to solve these convex problems.Moreover,we prove that the DCA method converges to a Karush-Kuhn-Tucker(KKT)optimal point under some mild conditions.Simulation results demonstrate that our algorithms can achieve significant performance gains compared with the existing algorithms.Secondly,we investigate the joint optimization of user association,subchannel allocation,and power allocation in multi-cell multi-association OFDMA heterogeneous networks.We obtain a locally optimal solution for the joint optimization problem by an alternating optimization method.To reduce the data exchange between the BSs to share the channel state information(CSI),we also consider the case that only statistical knowledge of inter-cell CSI is available.Then we extend the problem into the multi-antenna receiver case and the proportional fairness case,respectively.Our algorithm performs well when the user equipments apply multi-antenna receivers,and it helps achieve a good tradeoff between throughput and fairness when considering proportional fairness.Thirdly,we consider the robust beamforming design in a two-tier heterogeneous network.Our objective is to maximize the worst-case Signal-to-Interference-plusNoise Ratio(SINR)in the small cell user equipment(SUE)while guaranteeing that the interference from the small cell BS to the macro cell user equipment(MUE)is less than or equal to a threshold.Such a problem is nonconvex with an infinite number of constraints.Nevertheless,we first transform the problem into an equivalent one which is more tractable and then propose an efficient algorithm to obtain a near optimal solution.Simulation results are presented to validate the effectiveness of the proposed algorithm.Furthermore,we obtain an upper bound of the original problem.In summary,we investigate user association,subchannel allocation,power allocation and robust beamforming design in heterogeneous networks.We proposes various novel resource allocation algorithms in heterogeneous networks.These algorithms can achieve better performance compared to the existing algorithms.This thesis not only enriches the state-of-art results of the resource allocation in heterogeneous networks,but also provides guidelines for the practical system.