用于视频多播传输的压缩传感实现方法研究  

Method of Video Multicast Achieved Based on Compressed Sensing

作　　者：赵慧民[1,2] 郭一缜[2] 丁晓艳[2] 朱立[1] 

机构地区：[1]广东技术师范学院电子与信息学院,广东广州510665 [2]西安邮电学院通信与信息工程学院,陕西西安710121

出　　处：《中山大学学报（自然科学版）》2012年第1期45-49,共5页Acta Scientiarum Naturalium Universitatis Sunyatseni

基　　金：国家自然科学基金重点资助项目(60633030);国家自然科学基金资助项目(90604008);广东省自然科学基金团队资助项目(04205407)

摘　　要：在较高的包丢失率和噪声信道环境下,提高视频传输的图像质量是通信技术应用的关键。基于压缩传感理论,提出一种实现多播传输的视频图像解码方法。通过研究测量矩阵与稀疏基的最佳配置关系以及视频帧间相关性,实现了一种视频帧内重加权l1范数解码和基于运动矢量的帧间解码。在噪声和包丢失信道模型下,通过与软投影解码实验比较,说明了实现方法的有效性。With higher packages loss ratio channel and noise channel,improving image performance of video multicast transmission is an importance problem for communication technology.Based on compressed sensing,a decoding method of recover image was achieved as videos are suitable for multicast transmission channel.By investigating the best configuration of measurements matrix and sparse bases as well as video inter-frame correlation,the reweight l1-minimization decoding of video intra-frame and inter-frame decoding with motion estimation were realized.At random Gaussian noise and loss ratio channel model,comparing the method with Softcast,experiment results show that the method is effective.

关 键 词：视频多播 压缩传感 矩阵 解码 

分 类 号：TN911[电子电信—通信与信息系统]