Research On Buffer-based Wireless Resource Allocation In High-speed Rail Mobile Communications

With the rapid development of mobile communication and the wide application of intelligent devices,the high-speed rail wireless communication system has increasingly higher requirements for rate,delay and broadbandization.Different from the traditional cellular structure,the base stations(BSs)in the railway communication system are mainly built on both sides of the railway,and the coverage area is distributed in a strip.In order to avoid the data loss and communication delay of trains in the edge of the cell,the caching is introduced in the high-speed rail mobile communication system.In addition,as an important part of mobile communication technology,wireless resource management can improve system performance while ensuring service of quality(Qo S).Therefore,this thesis mainly focuses on the research of wireless resource allocation based on cache in highspeed rail mobile communications.The specific work is as follows:Firstly,this thesis deeply analyzes the related technologies of the high-speed rail mobile communication system.The characteristics of wireless channels are the key factors that affect the Qo S of communication,so this thesis first analyzes the channel fading characteristics and their signal transmission models.In order to improve the signal transmission quality and system performance,the mobile relay technology and distributed antenna systems(DAS)are applied to high-speed rail scenarios.Therefore,this thesis analyzes the advantages and disadvantages of mobile relay technology and DAS,and the applications in high-speed rail scenarios in detail.Then,this thesis also introduces the relevant knowledge of wireless resource management,in which we focus on the power control technology and the theoretical solution of wireless resource management problems,and its shortcomings in the high-speed railway scene.Secondly,this thesis proposes a power allocation method based on cache for uplink energy efficiency optimization of high-speed rail communication.In the proposed scheme,a mobile relay with data cache is installed on the top of the carriage,and the maximum system energy efficiency and power allocation method with buffer constraint are obtained by matching the data arrival and wireless channel service process from the access point(AP)of the train to BS.The optimization problem in this thesis can be modeled as a multi-constrained nonlinear fractional programming problem.Considering the objective function of fractional form and the non-convex constraint condition,the nonlinear fractional form problem is transformed by the auxiliary parameters,and the iterative algorithm is used to calculate the auxiliary parameters.Simulation results show that the energy efficiency of the proposed method is better than that of the traditional minimized power allocation method.Finally,this thesis studies the low complexity energy-efficient power allocation method for highspeed rail communication system assisted by DAS.In the proposed scheme,a data buffer area is equipped in the central unit,and the system energy efficiency is maximizied under the average minimum system data-rate constraint,the average maximum transmit power constraint and data buffer amount constraint.The joint power and rate allocation design with time-delay constraint is adopted to obtain the maximum system energy efficiency and low complexity optimal power allocation method.Because the optimization problem is a non-linear fractional form,it can be solved by parameter transformation and iteration method.However,for the multi-constrained fractional nonlinear optimization problem,the traditional solution method has a relatively high complexity,because it requires a two-layer iterative process,the auxiliary parameters need to be constantly updated in the outer iteration and the Lagrange multipliers need to be solved in the inner iteration.Then,this thesis classifies and discusses the final manifestation according to the compromise relationship between optimal energy efficiency and rate and power,and the feasible domain range of the optimal solution,and then uses the special update form of dichotomy to iterate and loop out the decision,thereby reducing iteration frequency.Theoretical analysis and simulation results show that this method can reduce the algorithm complexity without losing the energy efficiency of the system.