Interference Coordination And Resource Allocation For Full-Duplex Cellular Networks

In the coming era of big data,cloud computing and artificial intelligence,the demand for data has increased exponentially,which brings about great opportunities and challenges to wire-less communications.The crucial problem for the incoming 5G era is the shortage of spectrum resources.Therefore,full-duplex(FD)communications,as a candidate for 5G communication networks,can potentially double the spectral efficiency(SE)compared with the half-duplex(HD)communication by allowing devices to transmit and receive signals simultaneously on the same channel.Under the conditions of incumbent hardware devices,the primary obstacles to the FD communication are high residual self-interference(SI)and co-channel interference(CCI).There are two types of FD communication topology.The first one is the bi-directional communication topology mode,where both the base station(BS)and users work in the FD mode with simulta-neous transmission and receiving.The other one is the three-node topology mode,where users work in the HD mode due to the hardware limitation while the BS works in the FD mode with the help of self-interference cancellation(SIC)receivers.Since current user equipments(UE)do not have the ability to perform SIC,practical communication systems usually adopt the three-node topology mode.Fortunately,existing SIC receivers are able to suppress the self interference to the noise level,which makes FD communication applicable.In this paper,FD communication is first applied to a single-cell network.Considering the three-node topology mode,we formulate a joint user pairing,resource block allocation,and power control problem.Due to the existence of 0-1 variables,the problem is NP-hard.We first relax the discrete variables into continuous ones and then use the concave-convex procedure(CCCP)to find the upper bound of the original problem.In order to reduce the complexity of the algorithm,we also propose a heuristic algorithm consisting of three steps:user pairing,resource block allocation,and power control.Simulation results show that the performance of the heuristic algorithm is close to the CCCP algorithm,which is significantly better than the HD system.Since the FD communication is more suitable for deployment in small cell networks,we apply the FD communication to multi-cell heterogeneous networks.In a heterogeneous network,a macro BS is set in the center of a macro cell and operates in the HD mode,while randomly deployed small base stations are set in the center of their respective small cells with FD communication.In order to offload more users to small cells,we introduce the cell range extension(CRE)technology to better balance the load between macro cell and small cells.Owing to the FD communication,the inter-cell interference becomes more complicated.Therefore,we propose an inter-cell interference coordination scheme,including base station association,user pairing,resource block allocation and power control.This scheme effectively reduces the complex inter-cell interference and significantly increases system capacity compared with the HD system.Finally,we study the application of the FD communication in ultra-dense heterogeneous net-works(HetNet).In a ultra-dense HetNet,small base stations are densely deployed within a macro cell.The macro base station only provides network coverage rather than directly communicates with users.Each user is connected to a small cell base station.Because small cell base stations are ultra-densely deployed,their numbers are comparable to that of users.The application of the FD communication in such ultra-dense HetNets requires new resource allocation schemes to maximize the system performance.We propose a full-duplex HetNet resource allocation scheme,which in-cludes base station association,mode selection and user pairing,power control and resource block allocation.Given the tradeoff between system capacity and user fairness,we propose a resource block allocation problem based on Nash Bargain Solution(NBS)and use the branch-and-bound(B&B)method to achieve the optimal resource block allocation scheme.Finally,we elaborate on how the macro base station exchanges the signaling with the small base stations via the X2 interface in the proposed scheme.