Resource Allocation And Interference Management In Heterogeneous Networks Of Mobile Communications

A multi-tier heterogeneous networks (HetNets) architecture which consists of macrocells, different types of licensed small cells, cognitive radio (CR) networks, cooperative relay-based networks and device to device(D2D) communication networks is considered to be one of the key architectures for the future 5th Generation(5G) Mobile systems. This architecture can effectively improve the spectrum efficiency and throughput of wireless networks, expand the coverage of hot spots,reduce communication delay as well as ensure a good end to end performance. However,it is confronted with enormous challenges on resource allocation and inter-ference management. Besides, with the rapid development of the wireless communication industry, the huge energy consumption of wireless networks severely restricts the sustainable development of the environment.In recent years, green communications which is committed to improve the energy efficiency of the wireless network has received widespread consideration. On this basis, this dissertation investigates the resource allo-cation and interference management in HetNets from the perspective of green communication leveraged by means of optimization and game theory.The main contributions are summarized as follows.For the downlink communication by deploying femtocells in the two-tier HetNets, the femtocells may create "dead zones" for the macrocell users (MUs) due to spectrum reuse. To avoid deteriorating the quality of service (QoS) of MUs and restrict the cross-tier interference within a predefined threshold, we propose a power control scheme based on the interference pricing at Macrocell base station's (MBS's) side. A Stackelberge game (SG) model is used to model the power control scheme, in which the MBS is regarded as the leader and the femtocell base stations (FBSs) are regarded as the followers. Specifically, the MBS prices the interference from femtocells and benefits by selling the interference quota to the femtocells. Subsequently,the FBSs adjust their transmit power according to the price. We proof the existence and uniqueness of Nash equilibrium (NE)strategy for the game model and propose a distributed interference price and power update algorithm with lower communication overhead. Simulation results show that the proposed algorithm converges quickly and the throughput performance can approach that of a cooperative scheme.Based on the above system model, we further investigate the resource allocation and power control schemes in terms of energy efficiency-fairness and energy efficiency-optimality, respectively. For the energy efficiency fairness scheme, we view this scheme as to solve a max-min fractional programming of energy efficiency. This problem can be solved by using generalized Dinkelbach's algorithm (GDA). For the sub-problem in the GDA, combined with a principle of fairness, we propose a distributed algorithm with lower communication overhead. For the energy efficiency-optimality scheme, the original problem is hard to solve due to the existence of co-tier interference. Leveraged by means of the lower bound of logarithmic function,we transform the objective function of the original problem to a lower bound function and solve the trans-formed problem by Dinkelbach's algorithm. Then, we design a heuristic iterative algorithm to approach the solution to the original problem. Simulation results verify the feasibility and superiority of the proposed two schemes.In addition, Taking into account the priority of each smallcell, we further investigate the resource allo-cation and power control scheme for the weighted sum energy efficiency (WS-EE) maximization in smallcell networks,in which the weights accounts for the energy efficiency priorities for each smallcell. This is a mixed integer variables non-convex fractional programming and is known as NP-hard. Therefore, we decompose the problem into two subproblems and solve them subsequently. That is, given a feasible resource allocation strategy, we optimize the power control strategy firstly. Then, based on the resource allocation strategy, we optimize the power control strategy according to the resource allocation strategy. In particular, a pricing based distributed algorithm is proposed to solve the subproblem of power control at each smallcell base station' side(SBS's) side. The algorithm is strictly proofed to converge to a Karush-Kuhn-Tucker (KKT) point of the orig-inal problem given the resource allocation strategy. Simulation results show the performance of the proposed scheme outperforms non-cooperative scheme and a convex pricing scheme in terms of WS-EE.D2D can achieve the transmission of information bypass base station, reduce network latency and im-prove spectrum efficiency of network. However, the presence of D2D will also cause cross-tier interference to traditional network. Therefore, we investigate the power control and interference management problem in macrocell networks deploying D2D communications. In order to improve the overall spectrum efficien-cy of the networks, D2D communications reuse the uplink resource of the system. We also construct the problem as a SG model, in which the MBS is viewed as a leader to manage the interference by setting up a price for reusing its uplink resource, each D2D transmitter (D2DT) is deemed as follower to adaptively adjust its transmit power according to the price. Based on this game model, the scenarios that D2D pairs ffully cooperative and partially cooperative are considered herein to investigate the maximization of the sum rate of D2D communications and the sum rate of each D2D pair, respectively. In addition, the price and power update algorithms are also designed for both scenarios. Simulation results verify the feasibility of the proposed schemes and show the superiority over a non-cooperative scheme in terms of the sum rate of D2D communications.CR technology can let user equipments with cognitive ability access to the spectrum hole of traditional network to share licensed spectrum and thus improve spectrum efficiency of the network. In this paper, we designed a two-stage spectrum trading scheme for heterogeneous network deploying cognitive cells. The non-cooperative and cooperative modes are adopted to model the spectrum trading process. In particular,for the mode of non-cooperation, we analyze the existence and uniqueness of spectrum pricing and bidding strategy for multiple service providers and multiple cognitive users, respectively. A distributed strategy up-dating algorithm is proposed and the stability condition for the algorithm is also analyzed. We evaluate the performance of pricing and bidding strategy as well as compare the overall revenue of the system for both modes.