Research On Joint Resource Allocation Schemes And Efficient Caching Methods In Cloud Radio Access Network

The fifth generation mobile communication system(5G)puts forward higher requirements on spectrum efficiency(SE),energy efficiency(EE),network cost,transmission delay and other performance.In order to effectively cope with the above challenges,the 5G system uses the Cloud Radio Access Network(C-RAN)as a key candidate technology.In C-RAN,the baseband processing part is aggregated and shared in a virtual Base Band Unit Pool(BBU Pool),which can effectively adapt to non-uniform services(such as the tidal effect of data)and utilize resources(e.g.calculate ability of base station)more efficiently.Remote Radio Heads(RRHs),with limited function,preprocess and forward the signal received from mobile users to the BBU Pool for centralized processing,via wired/wireless fonthaul link.Compared with traditional communication architectures,C-RAN requires fewer BBUs and saves network cost by deploying and upgrading BBUs and RRHs flexibly due to actual network conditions.Meanwhile,a shared virtual BBU Pool could improve spectrum efficiency,reduce inter-cell interference and interaction delay of BSs,and achieve inte-cell load balancing.However,to facilitate the C-RAN technology,there are still many issues that need to be solved,such as,moderate energy efficiency improvement scheme and design under the quantization mechanisms of RRHs,joint resource allocation algorithm design in multi-objective confliction scenario,optimized RRHs selection algorithm design for throughout maximization,high energy-efficient resource allocation algorithm and caching scheme design in cache-based RRHs scenario,minimization of transmission delay design for in cache-based RRHs scenario.Based on these,we conduct the topic 'Research on Joint Resource Allocation Schemes and Effective Caching Methods in Cloud Radio Access Network',the main contributions are listed as follows:1.In an OFDMA-based C-RAN system,under two quantization schemes on the RRHs side,in order to improve the energy efficiency of users,a variety of joint resource allocation algorithms are proposed.Considering the fronthaul capacity and the users' uplink power constraint,the fractional programming is utilized to transform the non-convex problem,and the Lagrangian dual method is designed to design an iterative algorithm in the single-RRH uplink transmission scenario.In the multi-RRH uplink transmission scenario,the Modified Particle Swarm Optimization(M-PSO)algorithm is raised to solve the more complex non-convex problems,which avoids the inefficiency caused by the direct use of single-user solutions.Through effective algorithm designing,under the constraints of user transmit power and fronthaul capacity,the energy efficiency of users is maximized by jointly optimizing the uplink transmission power of users and fronthaul capacity of RRHs.The experimental results show that the proposed algorithms could effectively improve the energy efficiency of users in C-RAN and reduce the system power consumption,which meets the requirements of future green communication.2.The joint resource allocation problem of Multi-Objective Optimization(MOO)is investigated for the multi-objective coupling scenarios in uplink C-RAN system.In C-RAN,different requirements lead to different optimization goals and most of goals are coupled with each other,which cause the conflicting relevance such that improvements in one objective cause degradation in the other objectives.Therefore,the problem of MOO still exists in C-RAN.In the uplink transmission of multi-objective coupling scenario in OFDMA-based C-RAN system,based on Gaussian quantization scheme,the mathematical model of Fonthaul Efficiency(FE),Spectum Efficiency(SE)and Energy Efficiency(EE)is established respectively.Based on the above different objectives,the MOO optimization problem of FE-SE and EE-SE is also set up.Through the Weighted Sum(WS)method,the original complex non-convex MOO problem is transformed into simple Signal-Objective Optimization(SOO)problem,and the corresponding joint resource allocation algorithm is designed for single-user scenario and multi-user scenario severally.The simulation results indicate the proposed algorithms could balance the tradeoff between multiple objectives availably and furnish a new perspective for resource allocation in C-RANs,compared with considering SOO only.3.In C-RAN downlink transmission system,the issue of how to improve the performance of the system under fronthaul capacity constraint by optimizing the diversity gain of RRHs selection is stduied.In C-RAN,the fronthaul capacity,which means the capacity between BBU pool and RRHs,is limited.A common method of processing fronthaul capacity constraint is to quantize/compress signal at RRHs side,while the diversity gain brought by the RRHs selection is rarely considered to improve performance in literatures.As the number of active RRHs grows from zero,the total system throughput continues to increase when more and more RRHs are activated for data transmission.However,due to the fronthaul capacity constraint,the throughput stops growing when the number of active RRHs reaches a certain amount,so it is necessary to consider how to select the optimal number of active RRHs.Firstly,under different channel conditions,the throughput expression for the given number of RRHs is derived.Secondly,an algorithm which selects the optimal number of RRHs is designed to maximize the throughput,and the asymptotically optimal closedform expressions for the number of selected RRHs under low SNR and high SNR conditions is derived.Finally,simulation results show that the RRH selection scheme proposed in this paper can improve the throughput effectively.4.In the scenario of enhanced RRHs with caching,a joint optimization algorithm for variables such as subcarriers is designed for the problem of minimizing the power consumption of RRHs.In C-RAN,most of the existing work assumes RRHs couldn't cache content.To better adapt the content-centric feature of next-generation communication networks,it's necessary to consider caching function for RRHs in CRAN.In an DL OFDMA-based C-RAN system,a jointly optimization scheme of SC allocation,RRH selection,and transmission power to minimize the total downlink power consumption is proposed when a Lagrange dual decomposition is utilized to design the optimal allocation scheme to transform the original non-convex problem.The experimental results indicate that the proposed optimization algorithms could effectively improve the energy efficiency of the system and save system power.5.In the scenario of enhanced RRHs with caching,under the Hybrid Automatic Repeat re Quest(HARQ)retransmission mechanism,the mathematical model of transmission delay is established and the optimal cache allocation algorithm is designed for the minimization of transmission delay.Under the premise of the orthogonality between the fronthaul link and the download link spectrum channel,the system employs the Markov chain theory to establish the minimization optimization problem of system transmission delay,when considering the specific transmission mechanism and probability of successful packet transmission.Considering the optimization objective function expression could only be obtained recursively,the Brain Storm Optimization(BSO)algorithm is introduced to solve the non-convex problem and obtain the optimal caching scheme.The experimental results demonstrate the proposed algorithms efficaciously reduce the transmission delay of the system and improve the system performance.