Research Of Adaptive Transmission Of Mobile Edge Computing Based On Radio Network Information

With the continuous development of mobile Internet technology,the emergence of high-definition video,augmented reality and virtual reality applications have put forward high bandwidth and low latency requirements for mobile networks.According to Cisco forecasts,by 2020,global IP video traffic will account for 82%of all consumer Internet traffic.The proposed mobile edge computing is referred to as a key technology and architectural concept for the transition to 5G,with mobile edge computing emphasizing closer to users,reducing latency in network operations and service delivery.At the same time,mobile operators can open more network information to third-party developers,so that they can provide better applications and services to users.Based on the mobile edge computing to study the video transmission service,on the one hand,it effectively promotes the construction planning of the existing edge computing network architecture,and improves the network from the service perspective;on the other hand,optimizes the video transmission service through mobile edge computing to improve the user experience quality.This thesis is mainly engaged in the research of mobile edge computing adaptive transmission mechanism based on radio network information.Firstly,a radio network information service platform is built on the 4G LTE mobile edge computing platform,and a set of services and network-assisted video transmission are deployed on the platform.The thesis proposes related solutions for bandwidth management and video quality adaptation of edge networks.The main work of this thesis includes:1.Design and implementation of radio network information service platform in mobile edge computing.Based on the mobile edge computing platform under the 4G LTE network,combined with ETSI's white paper on MEC,a radio network information service platform capable of providing radio network information is built.Then,the radio network information service platform and its key technologies are elaborated in detail.Finally,the platform is tested.The test results show good performance of the radio network information platform.2.Research on adaptive video transmission based on radio network information to solve the fairness of bandwidth allocation.Firstly,the dynamic adaptive video streaming framework and implementation are introduced.Aiming at the fairness of video transmission in cellular networks,a solution combining radio network information of users is proposed.Dynamically adjust network resource allocation by detecting real-time load conditions of individual user links under cellular networks,and the design experiment compares with the existing client-based adaptive video streaming scheme.This mechanism can ensure the relative fairness of video quality viewed by users under the entire network.In the case of this experiment,the results confirmed a 32%improvement in the fairness of the scheme driven entirely by the client.3.Research on adaptive video transmission based on radio network information to increase QoE.Combining the MEC platform with radio information and the Q-learning algorithm,a QoE-oriented adaptive transmission algorithm based on radio network information is proposed.The algorithm uses the trained model to determine the quality of the video played according to the client status.Firstly,the feasibility and accuracy of bandwidth estimation using radio network information are verified by experiments.Secondly,compared with buffer-based and bandwidth-based adaptive algorithms,the results verify the effectiveness of the proposed scheme.In the case of this experiment,compared with the traditional BB A algorithm,the QoE indicator has increased by 18%.Experiments have confirmed the low latency of the MEC.