Research On Cooperative Management Technology Of Wireless And Optical Resources For Video Transmission

With the continuous development of 5G communication technology,the number of user terminal devices is increasing,and the proportion of mobile video data in global traffic increases sharply.Video service has gradually become the mainstream service of mobile communication Internet.At the same time,in order to meet the demand of massive video content transmission and achieve the technical goals of ultra-high capacity,ultra-low delay and ultra-large connection,it is great important to innovate the traditional network architecture and choose the appropriate fronthaul network hosted scheme.Time and wavelength division passive optical network is used in 5G access network to provide high-capacity and highreliability network access services for users.It can reduce deployment cost and operating cost and provide more flexible network expansion service,which is the application of a wide range of forward network solutions.However,there are few studies that considered the cross-domain resource allocation of wireless transmission and optical fronthaul network.This thesis focuses on the collaborative management technology of wireless and optical resources for video transmission,aiming at improving the resource utilization of wavelength capacity in 5G wireless domain communication resources and passive optical forward network,improving the user’s QoE and reducing the power cost of the system.Firstly,in view of the video unicast transmission scenario,based on the adaptive streaming media transmission architecture of wireless and optical fusion networks,this thesis proposes joint beamforming,video rate adaptive selection,and wireless and optical multi-dimensional resource allocation management mechanism.Firstly,the optimization of the proposed architecture design is discussed.Secondly,the established joint optimization problem is reconstructed,and the joint beamforming,rate adaptive selection and resource allocation algorithms were designed based on the convex-concave process optimization method.Compared with other benchmark schemes,a wireless and optical multi-dimensional resource cooperative management scheme for HTTP-based dynamic adaptive streaming video unicast transmission proposed in this thesis,has achieved at least 9.7%performance improvement in the overall utility of the system.Secondly,for video multicast transmission scenarios,user grouping mechanism is introduced in this thesis in addition to the multi-dimensional resource co-optimization,and an optimization scheme based on scalable video coding hierarchical transmission mode is designed,which jointly considered video selection at the user enhancement layer and resource allocation in wireless and optical networks.Firstly,the initial group of users is determined according to the user content request,and the optimization strategy of joint beamforming and wireless and optical multidimensional heterogeneous resource allocation is designed to maximize the overall utility of the system.Then,the user enhancement layer video is selected based on the last resource allocation results,and the user group is updated.By iteratively updating user groups and resource allocation,the optimal solution of the original problem is sought.The simulation results show that compared with the existing algorithms,the multi-dimensional resource joint allocation scheme proposed in this thesis improves the overall utility of system by at least 7%and achieves a better trade-off between different performance indicators.