Research On Resource Management For VR Services In Mobile Edge Networks

In recent years,with the development of network communication technology,the commercialization of 5th generation mobile networks(5G)and the upgrading of related equipments,virtual reality(VR)services and applications have been greatly developed.In the scenarios of VR video services,the amount of panoramic video data is usually several times more than that of traditional video,and because users' Quality of Experience(QoE)requires extremely low latency,the pressure on base stations in the network has increased significantly.In order to reduce the pressure on the base stations in the wireless network and meet the user's demands for panoramic videos,the introduction of Mobile Edge Computing(MEC)technology would sink the core network content distribution function to the network edge.Utilizing its data caching and computing capabilities,it can effectively reduce the latency and network load during panoramic video transmission.Therefore,in VR wireless networks,how to effectively use MEC's video caching and network communication transmission resources would become a key issue.Comparing to traditional videos,panoramic videos are multi-angle.Due to the visual characteristics of human eyes,it is impossible to watch full of the whole 360° video content at the same time.Therefore,in view of the multi-angle feature of panoramic video,transmitting the field of view(FOV)content can make better use of network resources and achieve load balancing.Based on the ultra-large transmission data,ultra-low latency requirements,and panoramic multi-angle characteristics of VR panoramic videos,this thesis focuses on the problem of network resource allocation algorithms in VR wireless networks.The main work of this thesis is as follows:Firstly,aiming at the system gain in VR wireless network,this thesis proposes a joint optimization of MEC caching placement,FOV control and channel allocation in a MEC-assisted VR wireless network to maximize utility function resource allocation.The resource allocation problem and the constraints of latency and channel have complex non convex expressions,which are difficult to solve directly.In this thesis,by deriving the mathematical properties of the constraint function,the latency and channel constraints in MEC assisted VR wireless network are solved through the difference of convex(DC)programming and variable relaxation.Finally,the convergence of the proposed resource allocation algorithm is verified by simulation,and the simulation results prove that the proposed algorithm has obvious gain compared with other schemes and can significantly improve the utility function of the system.Secondly,based on the system gain in the multi-cell cooperative scenario in the VR wireless network,this thesis proposes a joint optimization of MEC caching placement,FOV control and cooperative caching strategy in a MEC-assisted VR wireless network to maximize the system utility function.The resource allocation problem is a mixed-integer non-convex optimization problem,which cannot be solved directly.This thesis divides the problem into three sub-problems.In the first sub-problem,given the caching cooperation strategy and FOV control scheme,the optimal caching placement strategy is obtained through variable substitution and Taylor expansion;in the second sub-problem,given caching cooperation strategy and the result of sub-problem one,we can obtain the optimal FOV control scheme;in the third sub-problem,based on the output of the first two sub-problems,the caching cooperation strategy is obtained.Then,the three sub-problems are alternately iterated to obtain the maximum system utility function,and the corresponding resource allocation algorithm is proposed.Finally,the simulations verify the convergence of the proposed resource allocation algorithm.The results demonstrate the effectiveness of the proposed algorithm which significantly improve the gain of the system and play a load balancing effect.Finally,aiming at the energy consumption in VR wireless network,this thesis proposes a MEC-assisted VR wireless network to optimize MEC caching placement,FOV control and cooperative caching strategies to minimize energy consumption.In order to meet the demand of data transmission latency and network load in VR wireless network,the resource allocation based on caching is studied.In this thesis,the optimal resource allocation strategy is obtained through analysis.Finally,the simulations verify the convergence of the proposed resource allocation algorithm.And the results also prove the effectiveness of the proposed algorithm which significantly reduce the energy consumption of the system.