Research On Interference Mitigation And Beam Resource Allocation Method For 5G Mmwave Communication

To solve the severe challenges of Sub-6GHz spectrum resource shortage and network densification,the millimeter wave(mmWave)communication can support 5G new services,such as the 1Gbps high-speed transmission for the virtual reality(VR)service,and has been regarded as one of the key technologies for 5G communication.However,the network sum rate faces new challenges in terms of the differentiated backhaul capacities of small-cell base stations(SBS s)and the various kinds of interferences between different access links in the ultra-dense network(UDN).Furthermore,considering the user mobility,how to predict the mmWave inter-beam interferences dynamically and guarantee the efficient transmission of diverse service contents need to be solved urgently.Therefore,in order to improve the utilization of mmWave spectrum resources efficiently,the interference mitigation and beam resource allocation methods for the non-mobile and mobile scenarios are proposed in this thesis,respectively.The key contributions of this thesis for the mmWave beam resource allocation mainly includes two parts:(1)Aiming at the problem that the traffic load-balancing and the interference mitigation in the UDN scenario cannot be jointly optimized to improve the network sum rate,a network sum rate optimization algorithm between SBSs and users is proposed,which is based on the many-to-many matching,and only requires linear computational complexity.Furthermore,the mmWave transmit power optimization strategy is designed for SBSs based on the convex optimization theory,where both the upper and lower convex bounds of SBSs'power are theoretically derived.In the meanwhile,the approximated convex problem for the SBSs'transmit power allocation is proved to finally converge to the Karush-Kuhn-Tucker(KKT)point.Simulation verification shows that compared with the traditional algorithms,the proposed algorithm can improve the network sum rate by 71.9%,and the average successful transmission probability can reach 89.0%.(2)Aiming at the problem that the diverse VR services cannot meet the requirements of efficient content transmissions for dynamic vehicular network,an interference prediction assisted V2I(Vehicle to Infrastructue)vehicle selection algorithm is proposed,which can be aware of the intra-SBS interference dynamically.Thus,the cumulative sum rate of V2I communication and the number of vehicles'cached content fragments can be well increased.Besides,a coalition game based V2V content distribution algorithm is proposed to guarantee the linear complexity and improve the content distribution efficiency between the transmitting vehicles and the receiving vehicles.Simulation verification shows that the average successful transmission probability of the proposed V2I vehicle selection algorithm and the V2V content distribution algorithm can still be as high as 72.63%,and can reach up to 53.4%gain over traditional algorithms.