项目名称: 网络信源编码的率失真理论和应用
项目编号: No.61271221
项目类型: 面上项目
立项/批准年度: 2013
项目学科: 无线电电子学、电信技术
项目作者: 王嘉
作者单位: 上海交通大学
项目金额: 75万元
中文摘要: 网络信息论是现代信息论的重要组成部分,网络信源编码是其中不可或缺的重要支柱。传统方法已经不能适应网络信息系统,网络信源编码的率失真理论已成为亟待发展的问题。在网络信源编码中,直接/间接多终端编码是核心和基础问题,也是信息论发展几十年来的著名未决难题,已成为网络信源编码理论发展的瓶颈。针对标量信源的传统方法已不再适用于矢量信源。本项目在申请者前期研究成果的基础上,以估计理论为基础,将多种信息论不等式融合,建立信息论极值不等式的统一理论框架;解决或部分解决针对矢量信源的多终端信源编码率失真区域的刻画问题;并进一步向复杂网络信源编码演进,建立一般编码器/解码器网络的信源编码率失真理论。本项目将通过一系列的方法和途径,在网络信源编码理论的传统难题上取得突破,从而推动网络信息论的发展。同时设计并实现面向特定网络的视频信源编码,为网络信源编码的视频应用提供借鉴。
中文关键词: 极值不等式;多终端信源编码;率失真理论;;
英文摘要: Network information theory is an important part of modern information theory, with network source coding being its fundamental. Traditional method is no longer suitable for the development of the network system, and the rate distortion theory of network source coding is an important problem to be solved. In network source coding, direct/indirect multiterminal source coding is the core and remains open after several decades. The multiterminal source coding problem has become the bottleneck of the network information theory, and traditional method used for scalar source is no longer available. Building on previous results in closely related areas, this project will establish a unified framework for information extremal inequalities by connecting the estimation theory and different kinds of information inequalities. The description of the rate distortion region of the direct/indirect multiterminal source coding problem is to be solved in this project, completely or partially. This theory will be further generated to more complicated network, and a general rate distortion theory will be established for general encoder-decoder network. Accordingly, this project will make breakthroughs in these fundamental problems in network information theory and promote the development of network information theory. And at the mean
英文关键词: Extremal inequality;Multiterminal source coding;rate-distortion theory;;