Research On Multi-antenna Transmission Technique Based On Cloud Radio Access Network Architecture

Faced with the growing demand for new mobile Internet applications in terms of data rate,network capacity and energy efficiency,technologies in the existing wireless network architecture are no longer sufficient to meet the above challenges.The centralized deployment of the network based on the cloud radio access network architecture can reduce the operation cost and significantly improve system architecture,mobility,coverage and energy efficiency,but it is also limited by conditions such as fronthaul capacity and transmit power.However,the existing common physical frequency bands are too crowded to meet the requirements of high quality wireless communication.Thus,people turn their attention to the millimeter wave band(30GHz-300GHz)that still has wide free range.For the multi-antenna transmission technique in cloud radio access network architecture,this paper studies the robust beamforming optimization of cloud radio access network(C-RAN),the distributed precoding optimization and the joint design of remote radio head(RRH)selection and hybrid precoding optimization in a system combining of C-RAN and millimeter wave.The main work of this paper is as follows:Firstly,according to the characteristics of centralized deployment,two precoding implementation methods,data sharing and compression strategy,and typical multicast precoding in C-RAN are studied.The busy signal exchange between the BBU and the RRH and the capacity limitation of the fronthaul link make the channel estimation error,distributed precoding and RRH selection a problem to be considered when designing C-RAN precoding.Secondly,aiming at the problem of channel estimation error in the downlink of multigroup multicast C-RAN,an optimization problem that maximizes the sum of minimum user rate in each multicast group is studied.The original optimization problem is transformed into a solvable convex problem by semi-determined relaxation(SDR),successive convex approximation and S-lemma.Then,a robust optimization algorithm is proposed for this problem.Through simulation analysis,the proposed algorithm is greatly improved compared with the traditional non-robust design in terms of the sum rate performance.Thirdly,the fully distributed precoding algorithm in the cached radio access network architecture is studied when the signal exchange between cooperative radio access nodes does not exsit.The fully distributed optimization method means not considering the signal exchange between cooperative radio access nodes.First,for the architecture of ultra-dense RAN and the assumption that there is no direct link between each radio access node,the optimization problem of user weighted sum rate maximization(WSRMax)is constructed.Then,precoding for each radio access node is independently designed by defining a virtual WSRMax problem.The simulation analysis shows that the average available sum rate performance of the fully distributed algorithm is close to that of the centralized optimization algorithm.Finally,the problem of joint RRH selection and hybrid precoding optimization in the millimeter-wave C-RAN architecture is studied.The problem includes mutually coupled optimization variables and non-convex constraints,so a hierarchical iterative algorithm is proposed.First,the joint optimization of the digital precoding and quantization noise,and the optimization of the analog precoding in the inner layer are performed.Then,the hybrid precoding iterative optimization of the outer layer is performed.The simulation analysis shows that the proposed hybrid precoding algorithm can converge quickly and achieve a balance between the system's sum rate and total power consumption,and achieve the goal of maximizing system energy efficiency.