项目名称: 基于合作博弈论优化的多视点信源协作编码与传输
项目编号: No.61472234
项目类型: 面上项目
立项/批准年度: 2015
项目学科: 计算机科学学科
项目作者: 邹君妮
作者单位: 上海交通大学
项目金额: 82万元
中文摘要: 为适应日益发展的无线协作通信网对多视点信源的交互内容服务,探索在动态系统约束中的联合表示和优化性能。依据多视点渐进分层编码,设计鲁棒和稳定的合作策略,确定动态网络节点间的约束性描述,进行合作博弈的传输优化。建立多视点信源间资源的分配公平性,结合多视相关信源的选择性、网络的动态随机性,建立动态优化的效用最大化(NUM)问题。采用多尺度几何的联合约束,进行边信息的协同描述和多尺度几何变换编码,刻画多维拓扑空间的奇异性;在此拓扑映射的视点基础上,结合时-空分级结构,优化数据层的分级调度;通过该多信源的约束,实现基于机会路由与马尔可夫信道模型的联合资源优化,进而基于纳什议价实现无线中继协作网络的帕累托最优功率配置,逼近动态网络的理论性能,建立多视点信源的编码与传输系统平台。本项目从实际应用中抽象出科学问题,涉及计算机视角、信号处理、信息论基础,预期的理论和技术成果可有助于交互多视点媒体应用的发展。
中文关键词: 多视点信源;合作博弈论;分层编码;机会路由;组播
英文摘要: To match the increasingly development of interactive multiview video dissemination over promising wireless cooperative network, we aim to investigate joint representation and optimization with stochastic constraints. With the progressive structure of multiview sources, it is dedicated to designing the stable strategy on cooperation and determing the coalitional form in cooperation game theory. To be concrete, it is formalized as a network utility maximization (NUM) problem in a dynamic optimization manner in terms of fairness, selectiviness, random accessibiltiy within multiview sources. It is achieved by advanced depth coding with multiscale geometric transform, optimized scheduling along the view-spatio-temporal resolution, resource allocation with opportunistic routing and dynamic Markov channel modeling, and power allocation in multi-source multi-relay newtork with Nash bargaining solution. The overall rate-distortion performance of multiview video coding is desired to be close to the theoretical bound of nonlinear approximation by the fusion of multi-dimension side-information. It is founded on the combinatorial area of computer vision, information theory and signal processing, and the achievements could favor the ongoing three-dimensional interactive media.
英文关键词: Multiview sources;cooperative game theory;layered coding;opportunistic routing;multicast