Research On Energy Saving Mechanism In Fiber Wireless Access Network With QoS-Aware Of Offloading

Fiber-wireless access network,as one of the important solutions of the next generation access network,has the high mobility,flexibility and high bandwidth and reliability of the optical network.Statistics show that in the 5G era,the massive number of intelligent access devices and the surge in device demand for high-speed and low-latency applications have caused excessive energy consumption for devices and access networks.Mobile edge computing offloading can effectively reduce energy consumption through the migration of computing tasks,but how to control the communication overhead of offloading and reduce the energy consumption of the access network on the basis of ensuring network performance and offloading quality is a key issue.Therefore,it is of great significance to design a reasonable offloading energy-saving mechanism.Firstly,this thesis describes the background,architecture,development prospects,and key technologies of fiber wireless access networks with edge computing offloading,and analyzes the problems of existing energy-saving mechanisms.Including optical domain energy saving,wireless domain energy saving and fiber-wireless integration energy saving,and propose solutions to existing problems.Secondly,this thesis proposes a fiber-wireless network energy-saving mechanism with offloading compression incentive.According to the different types of service attributes and the maximum tolerant delay,combined with the load changes of the optical network unit and the traffic of the wireless mesh network,dynamically adjust the ratio of the offloading compression of service to reduce the communication overhead of the offloading by the average arrival rate of the offloaded data of different priorities obtained by means of statistical methods and combined with the delay of compression of each node.At the same time,a queuing model is established to analyze the delay of the offloading service in the MEC server and cooperatively schedule the relay node in wireless mesh network,thereby performing the schedule of collaborative sleeping on the optical network units and the terminal devices to improve the energy efficiency of system.The results show that the proposed mechanism can effectively reduce the network energy consumption while ensuring the delay performance of offloading service.Thirdly,in order to solve the communication overhead and edge server deployment problems in the MEC offloading process,a cluster-based MEC deployment joint offloading compression mechanism is proposed.First,the multiple connection characteristics of the candidate nodes are analyzed to determine the cluster head node,and then the cluster member nodes are determined to complete clustering by the maximum hop limit in the cluster and the load constraints of the cluster head node.The nodes in the cluster forward the offloaded services to the cluster head node based on the service priority and link status,and deploy a MEC server to perform offload compression processing.Because the edge node transfers part of the load to the cluster head node through the clustering mechanism,the low-load node can be combined with the optical network unit to sleep to reduce energy consumption.Finally,the performance of the proposed algorithm is verified by simulation experiments.The results show that the proposed method can effectively reduce the energy consumption of the entire network while ensuring the delay performance of offloaded services.Finally,the full thesis is summarized,and future research work is expected.