多视点视频无线资源优化

多视点视频是指在场景中放置多台摄像机,记录下多个视点数据,提供给用户视点选择和场景漫游的交互式媒体应用.多个摄像机从不同视角同时拍摄同一场景得到的一组视频信号,每一个摄像机代表一个不同的视角.可同时传输多个空间角度的视频流到用户端,并合成用户所需要的视域图像.多视点视频是一种新型的具有立体感和交互操作功能的视频,是未来一种极具应用前景的多媒体应用.然而,当前多视点的无线网络带宽分配机制中,都没有考虑大量的、不同解码能力的用户共存时的效率问题.一般情况下,合成视域往往需要至少左右两边(两条以上)的参考视频同时传输到用户端,才能使合成的视域质量不低于直接传输的视点质量,使得网络数据量成倍增加.同时,用户的设备性能影响用户感知质量.网络中手机屏幕和高清大屏显示对网络传输视频数据率的要求不同,必须考虑用户设备解码能力的限制,才能真正提供用户满意的感知质量.本文通过考虑移动无线网络带宽约束,考察不同用户端硬件的解码能力、视域大小及带宽消耗,利用博弈理论,使整体网络资源收益最大.本文分别考虑了几种特定场景下多视点视频传输的无线网络的资源分配.第一,已知网络用户的满意度参数,不考虑带宽的限制(带宽充足),如何确定每个用户需要支付的单位价格.第二,在用户个数不确定的条件下,如何判断出可以接入的用户个数和用户需提供的单位价格.第三,同时考虑用户设备对最大场景复杂度解码能力受限以及网络带宽受限两个约束条件,同时进行用户接入控制和多视点的视频质量优化,使得网络的收益和用户的效用得到最大化.本文对提出的算法进行了理论分析,证明了本文参数设置的合理性.在多视点移动网络资源调度中,本文提出的算法可以方便设置所需的价格参数.从视域大小、价格、用户效用、网络收益等各方面对实验性能进行比较.仿真结果显示本方法在同等实验条件下,多视点用户效用提升分别为5%和12%,网络总体收益增加32%.本文算法可以同时满足网络收益和用户整体效用最优,提高多视点视频在多用户下的网络资源利用率.

Multi-View Video Wireless Resource Optimization

Multi-view video refers to placing multiple cameras in the scene,recording multiple view data,and providing the user with the interactive ability of view selection and scene roaming.A set of video signals obtained by multiple cameras simultaneously shooting the same scene from different perspectives,each camera represents a different perspective.Video from multiple spatial angles can be simultaneously transmitted to the client,and the image of view area required by the user can be synthesized.Multi-view video is a new type of video with three-dimensional sense and interactive operation function.It is a kind of multimedia application with great application prospect in the future.However,in the current multi-view wireless network bandwidth allocation mechanism,the efficiency of a large number of users with different decoding capabilities is not considered.In general,composite view area requires a higher transmission rate.Because the synthetic view area often needs at least left and right sides(two or more)of reference video to be transmitted to the client at the same time,in order to make the quality of the synthesized view area not lower than the quality of the directly transmitted viewpoint,making the amount of network data multiply.At the same time,the user’s device performance affects the user’s perceived quality.In the network,mobile phone users and high-definition large-screen display have different requirements on the data rate of video data transmission on the network.The increase of the rate can improve the quality of video to a certain extent,but the limitation of decoding ability of user devices must be considered to truly provide users with satisfactory perceived quality.In this paper,by considering the bandwidth constraint of mobile wireless network,the decoding capability,view size and bandwidth consumption of different client hardware are investigated,and game theory is used to achieve the maximum benefit of the whole network resources.The resource allocation for multi-view video in transmission wireless network is considered in this paper.First,if the satisfaction parameters of network users are known,how to determine the unit price needed to pay per each user,regardless of the bandwidth limitation(bandwidth is sufficient).Second,under the condition of the uncertain number of users,how to determine the number of users that can access and the unit price needed to provide per user.Thirdly,considering the limitation of decoding ability of user equipment to the maximum scene complexity and the limitation of network bandwidth,the user access control and multi-view video quality optimization are carried out simultaneously to maximize the benefits of the network and the utility of users.In this paper,the proposed algorithm is analyzed theoretically and the rationality of parameter setting is proved.In multi-view mobile network resource scheduling,the algorithm proposed in this paper can easily set the required price parameters.The experimental performance was compared from the aspects of view size,price,user utility and network benefit.Simulation results show that under the same experimental conditions,the utility of multi-view users was increased by 5% and 12% respectively,and the overall network revenue was increased by 32%.The algorithm in this paper can simultaneously satisfy the optimization of network revenue and the overall utility of users and improve the utilization rate of network resources in multi-view video with multi-users.