| With the popularity of smart phones,tablets and other smart mobile terminals,as well as the application of multimedia services such as video services,the number of mobile terminals and data traffic in mobile network are exponentially increasing.The traditional 4G network architecture cannot satisfy the dense service requirements of 5G.In order to solve this problem,a flexible radio access network architecture should be proposed.Cloud radio access network is considered as a novel radio access network architecture,which utilizes cloud computing and virtualization technology to migrate in-formation processing functions from the base stations to the baseband units pool.Mean-while,the baseband units pool is connected to the remote radio heads through fronthaul link.Cloud radio access network has the characteristics of distributed deployment and centralized processing,which can effectively improve the spectrum efficiency and energy efficiency of mobile communication system.Although cloud radio access network has many advantages,there are still some shortcomings such as large data transmission delay and limited fronthaul link capacity.In order to solve these problems,fog radio access network is proposed by migrating a part of computing resources from the cloud to the edge of the network."Cloud-fog" radio access network has flexible networking mode and efficient resource utilization,which provides a feasible solution to meet the dense service requirements of 5G.The purposes of "cloud-fog" radio access network are to improve system spectral efficiency,reduce energy consumption and guarantee service delay.In order to improve the system spectral efficiency,many effective solutions have been proposed.This thesis mainly studies the purposes of reducing energy consumption and guaranteeing service delay in "cloud-fog" radio access network.In order to reduce energy consumption,chap-ter 2 considers the tidal effect of mobile network,which will lead to dynamic changes of service requirements.A flexible resource allocation strategy is proposed to reduce ener-gy consumption.In order to guarantee service delay,chapter 3 considers the separation of information processing and transmission in cloud radio access network,which makes end-to-end delay include baseband units pool processing delay and wireless transmission delay.We propose a joint radio and computation resource allocation strategy to maximize the average throughput utility while guaranteeing the end-to-end delay in cloud radio ac-cess network.Chapter 2 and 3 are studied in traditional cellular network.In high-speed mobile scenario,because the moving direction of the vehicle is determined and the speed is known,the trajectory of the vehicle can be predicted.These characteristics can fa-cilitate resource allocation.At the same time,the high speed mobility of vehicle leads to time-varying channel,which brings difficulties to the resource allocation.In order to guarantee service delay in high-speed mobile scenario,chapter 4 considers the train trajectory can be predicted in high-speed railway scenario,which makes the predictable path loss information can instead of accurate channel state information.By utilizing this characteristic,a hybrid power allocation scheme is proposed to achieve the tradeoff be-tween energy efficiency and capacity.Chapter 5 considers the time-varying channel in vehicular network,which will effect the resource allocation strategy.Considering this characteristic,a joint radio and computation resource allocation strategy is proposed to improve the system utility in fog radio access network.The main contributions and innovations of this dissertation are listed as follows:1)In order to reduce energy consumption in traditional cellular network,considering the dynamic changes of service requirements caused by tide effect,we formulate the problem of joint ultra dense cells management and resource allocation,with the ob-jective to minimize the network power consumption while guaranteeing the quality of service requirements of users and the coverage rate requirement.The problem is for-mulated as a mixed-integer nonlinear programming problem.To deal with this prob-lem,we utilize the sparse characteristics of the beamforming vector and the method of reweighted l1 norm to approximate l0 norm.We propose an iterative algorithm to derive the lower bound of the problem,and then propose a heuristic iterative algorithm to obtain a practical solution.Simulation results confirm that the performance of the proposed heuristic algorithm can reduce energy consumption effectively.2)In order to guarantee service delay in traditional cellular network,considering the separation of information processing and transmission in cloud radio access network,which makes end-to-end delay include baseband units pool processing delay and wire-less transmission delay.We investigate the stochastic optimization problem to maxi-mize the average throughput utility while guaranteeing the end-to-end delay of cloud radio access network.By utilizing Lyapunov optimization technique,the primal prob-lem can be decomposed into four subproblems.By utilizing the results of four sub-problems,a joint admission control and resource allocation algorithm is proposed.The simulation results are provided to show that the complexity of the proposed al-gorithm is low,and it can achieve the tradeoff between the average throughput utility and queue backlogs.3)In order to guarantee service delay in high-speed railway scenario,considering the train location can be predicted,we investigate the tradeoff between energy efficiency and capacity with service delay requirements.Two problems are formulated,first,we consider that the communication system of high-speed railway only bears the passen-ger information services.The energy-efficient power allocation problem with delay constraint is studied.Second,we consider that the passenger information services and the train control services are all bore.A capacity optimization problem with joint energy efficiency performance and services transmission delay constraints is formu-lated.Based on the train location can be predicted,we propose two hybrid power allocation schemes to solve the two problems.The effect of small-scale fading on energy efficiency performance is analyzed.The effectiveness of the proposed power allocation algorithm is validated by high-speed railway channel measurement trace based emulation results and extensive simulation results.4)In order to guarantee service delay in vehicular network,we consider the problem of joint radio and computation resource allocation with the objective to maximize the system utility while guaranteeing the task delay requirements,in which the influence of time-varying channel on the task resource allocation strategies is considered.Due to the influence of time-varying channel,the task transmission time is related to the lo-cation of the vehicle and the allocated bandwidth,which makes the tasks transmission time uncertain.In order to deal with it,we first utilize the fixed spectrum efficiency(SE)instead of the time-varying SE,then propose a linearization based Branch and Bound algorithm to solve the fixed SE problem.Considering the complexity of the Branch and Bound algorithm cannot be guaranteed,a closest rounding integer algo-rithm is proposed to solve the fixed SE problem.After that,based on the resource allocation strategies of the fixed SE problem,we propose two heuristic algorithms to solve the original problem.Furthermore,we analyze the effect of small-scale fading on the proposed resource allocation strategies.The simulation results are provided to show that the proposed two heuristic algorithms can enhance the system utility effectively. |
PDF Full Text Request