Research On User-aware Video Transmission Strategy In FiWi Network

Fiber-Wireless Access Network is a new type of converged network that combines optical and wireless access networks.Fi Wi can provide users with high-capacity,highbandwidth,and anytime,anywhere network access experience.It is one of the important solutions to solve the "last mile" high-speed access network.Although the architecture of the Fi Wi access network provides good physical conditions for the transmission of video services,the transmission capacity between the optical domain network and the wireless domain network is too different.This will not only affect the transmission performance of the network,but also bring users a poor quality of experience.According to the characteristics of the network architecture and the user's experience requirements,designing a reasonable video transmission mechanism is essential to improve network performance and user experience quality.Therefore,based on the characteristics of the Fi Wi network architecture,this thesis propose a video transmission mechanism with Qo E constraints and load balancing and bandwidth allocation mechanism with user experience quality aware for the purpose of ensuring network performance and improving user experience quality.Aiming at the differentiated needs of users and the problem of interruption of video services caused by partial overload of the converged network,this thesis proposes a multipath video transmission mechanism with Qo E constraints.The path selection model is established through the front-end structural characteristics and link load information of the optical fiber wireless access network;set the threshold of video transmission delay and path differential delay on wireless side as user Qo E constraints;considering the user experience quality,the particle swarm optimization algorithm with multi-level penalty function is used for video distribution.Simulation results show that the proposed mechanism can effectively maintain network load balance and alleviate network congestion while avoiding user loss.For the transmission of real-time scalable video to multiple users in a wireless timevarying channel,bandwidth allocation plays an important role in the quality of user experience,and channel fluctuations bring great challenges to satisfying the quality of experience required by users.According to the video content information and video quality layer,the diffusion approximation method is used to establish the video playback diffusion process and the underflow probability model is derived;Then,the Qo E evaluation model is established based on the buffer underflow probability,the played video quality,and the utility function of bandwidth changes;Furthermore,the Lagrangian multiplier method is used to adaptively allocate the request bandwidth of each user to ensure the viewing experience of the entire user.Simulation results show that the proposed mechanism can avoid the interruption of video playback to a certain extent and improve the user experience quality.