Performance Optimization For Multi-base-station Cooperation Communication Systems

With the rapid development of mobile Internet,the number of various personal and commercial wireless equipment grows explosively.Therefore,the future mobile net-works will face the challenges of system capacity,reliability and low latency.To cope with these challenges,the researchers have put forward strict requirements on the com-munication performance of 5G networks and future super 5G as well as 6G networks.To meet these requirements,dense network and heterogeneous network deployment is required in the architecture of the future networks.However,this can bring severe inter-cell interference.Cooperation communication is an important approach for addressing the complex interference,which is also regarded as a key technology in the implemen-tation of the future networks.This dissertation focuses on multi-base-station cooperation communication sys-tems.Considering the density and heterogeneity of the future networks and the different transmittability and synchronization level between different scale of cooperation,this dissertation studies the MIMO(Multiple Input Multiple Output)cooperation in dense networks as well as the power control and the cooperation beamforming in general het-erogeneous networks.The contributions and novelty of this dissertation can be summa-rized as follows.Under dense network deployment,this dissertation focuses on the beamforming in MIMO cooperation communications.The total transmission power is optimized subject to user-specific SINR(Signal to Interference plus Noise Ratio)constraints.With the in-crease of scale(numbers of base stations,devices and antennas)of future networks,the scale of corresponding optimization problems is getting larger.In order to make full use of the computation resource of the cooperation base stations and reduce execution time of the beamforming algorithm,this work considers the distributed beamforming algo-rithm based on SCA(Successive Convex Approximation)and dual-decomposition.The signalling overhead of this algorithm is independent with number of antennas,which makes it very suitable for multi-antenna systems.Simulation results show that the exe-cution time of the proposed algorithm is much less than the centralized one.Compare to the ZF(Zeor Forcing)precoding and classic SOCP(Second Order Cone Programming)algorithm,the proposed algorithm can achieve higher energy efficiency.Under the power control,this work optimizes the total transmission power subject to user-specific QoS(Quality of Service)constraints.Traditional instantaneous perfor-mance based power control scheme is greatly affected by channel fluctuation.To avoid this deficiency,this work classifies the users into two categories according to their sen-sitivity to latency:DSUs(Delay-Sensitive-Users)and NDSUs(Non-Delay-Sensitive-Users).This work then utilizes instantaneous SINR constraints to ensure the success of data transmission per time slot to meet DSUs'low latency requirements,and the long-term mean data rate constraints to ensure NDSUs' average data rate requirements.Because of the mean data rate constraints,the power control problem is formulated as a non-convex stochastic constrained problem.The recently proposed CSSCA(Con-strained Stochastic Successive Convex Approximation)technique is utilized to handle this problem.The computation complexity and signalling overhead of the proposed scheme is also given.Simulation results show that the proposed scheme can signifi-cantly improve the feasible probability of DSUs' instantaneous constraints and reduce the network's energy consumption.Additionally,the proposed scheme demonstrates good scalability,which makes it applicable to large scale networks.Improvements to the traditional systems are difficult to fully meet the demand-ing performance requirements of future networks,so the IRS(Intelligent Reflecting Surface)is introduced into wireless communication systems recently.The IRS is com-posed of a large array of scattering elements,which can be individually configured to generate additional phase shifts to the signal reflections.Hence,it can actively control the signal propagation properties in favor of signal reception,and thus realize the notion of a smart radio environment.As such,the IRS's phase control combined with the con-ventional transmission control can potentially bring performance gain compared to the conventional wireless networks without using the IRS.With the aid of IRS,we study the cooperation beamforming of MIMO heterogeneous network and optimize the weighted sum rate.Unfortunately,this problem is non-convex and base station's transmission beams,user's receive beams and IRS's phase shift matrices are coupled together by the complicated data rate expression.Utilizing the alternative optimization method,we achieve the design of transmission beam,receive beam and phase shift matrix.How-ever,the objective of the IRS optimizing problem is a non-convex composite function because of the reflectivity-phase shift response.A SCA based tow-layer optimization method is proposed to handle this difficulty.Finally,the transmission beam,receive beam and phase shift matrix are modified according to the finite precision phase shift.Simulation results show that IRS can significantly improve the total data rate and the performance improvement is proportional to the number of IRSs.