Enhanced Interference Management Techniques for Heterogeneous Cellular Networks

Interference management is one of the most challenging problems facing wireless communication networks, especially for the cellular wireless communication system that is based on reuse-one deployment.
This problem becomes even more noteworthy in the heterogeneous cellular networks (HetNets) where lower power nodes (LPNs) are deployed in the coverage area of the macro base station (MBS). The higher transmit power possessed by the MBS, together with the cell selection procedure employed in HetNet: where a user equipment (UE) may be served by a closer LPN (to enable cell splitting) even though the received power from the MBS could be higher, are some factors that cause interference in HetNet. In the 5th generation mobile networks (5G) when the number of deployed LPNs increases interference will be more serious.
This thesis proposes interference management techniques based on beamforming with different level of cooperation amongst base stations in HetNet. In this thesis, we first designed global beamforming vectors that will maximize the weighted sum-rate of HetNet while fulfilling some power and interference constraints. The interference constraint controls the allowable interference from the MBS to other UEs in the HetNet. The global beamforming vectors were achieved using the Branch and Bound technique which is a global optimization method used in solving non-convex optimization problems. The beamformers that maximize the weighted sum-rate of HetNet are designed jointly by all BSs in the HetNet, hence the implementation is done centrally.
Since each UE in HetNet has peculiar interference situation, we design a UE-centric clustering scheme, which is capable of determining the BSs in the HetNet that interferes each UE the most at a particular time. Afterward, these BSs coordinate interference with the serving BS of this UE and make resource allocation decisions together to allocate beamforming directions and powers to each UE in the HetNet.
This will spatially separate signals sent to each UE, thereby mitigating interference and improving the total data rate achievable in HetNet.
HetNet tends to be distributed, also X2-interface which is the backhaul link that connects BSs in the HetNet has a limited capacity which makes it incapable of withstanding huge burdens in its backhaul. We,
therefore, design distributed beamforming directions using only local channel state information available at each transmitter. We also develop optimal power allocation scheme for each UE in each cell to maximize the sum-rate of each cell in the HetNet.