首页 | 官方网站   微博 | 高级检索  
     

视频处理器软硬件协同设计
引用本文:俞国军,刘鹏,姚庆栋,蒋志迪,蔡卫光.视频处理器软硬件协同设计[J].浙江大学学报(自然科学版 ),2006,40(7):1117-1122.
作者姓名:俞国军  刘鹏  姚庆栋  蒋志迪  蔡卫光
作者单位:俞国军,刘鹏,姚庆栋,蒋志迪,蔡卫光( 浙江大学 信息与电子工程学系,浙江 杭州 310027)
基金项目:高比容电子铝箔的研究开发与应用项目;霍英东教育基金
摘    要:为了提高视频图像处理速度与硬件资源利用,针对一种基于精简指令集处理器与数字信号处理器(RISC/DSP)混合体系结构的媒体处理器:浙大数芯(MD32),给出了一种软硬件协同设计策略.所给策略结合视频处理核心算法,研究分析MPEG视频编码标准的处理过程,进行了视频处理指令扩展设计,提高了数据的并行处理能力,利用了指令内并行执行特性.为有效实现扩展指令,处理器执行级采用了可扩展流水级技术.实验结果表明,指令扩展硬件成本仅占MD32的2.7%,逆离散余弦变换实现性能比MMX/SSE指令集实现的性能分别提高31%和23%,运动补偿性能比MMX指令集实现的性能提高了40%.

关 键 词:视频处理器  " target="_blank">lang="EN-US">  视频压缩  " target="_blank">lang="EN-US">  协同设计  " target="_blank">lang="EN-US">  单指令多数据
文章编号:1008-973X(2006)07-1117-06
收稿时间:2005-04-24
修稿时间:2005年4月24日

Video processor design using hardware and software co-design strategy
YU Guo-jun,LIU Peng,YAO Qing-dong,JIANG Zhi-di,CAI Wei-guang.Video processor design using hardware and software co-design strategy[J].Journal of Zhejiang University(Engineering Science),2006,40(7):1117-1122.
Authors:YU Guo-jun  LIU Peng  YAO Qing-dong  JIANG Zhi-di  CAI Wei-guang
Affiliation:Department of Information Science and Electronic Engineering, Zhejiang University, Hangzhou 310027, China
Abstract:A hardware/software co-design method for video processing instruction extension based on microprocessor MediaDSP3201(briefly,MD32) with unifying architecture of reduced instruction set computer and digital signal processor(RISC/DSP) was introduced to improve video image processing speed and utilization of hardware resource.The key algorithm in video processing was analyzed based on MPEG video coding standard.The proposed method improved the parallelism for data processing and utilized parallel processing characteristics within instructions.In the instruction execution pipeline stage,scalable pipeline technology was adopted to realize the video processing instruction.Results showed that hardware cost for the instruction extension was 2.7%,and that performance improvement achieved was 31% and 23% respectively compared to MMX/SSE for inversed discrete cosine transform(IDCT) and 40% for motion compensation(MC) compared to MMX.
Keywords:video processor  video compression  co-design  single instruction multiple data(SIMD)
本文献已被 CNKI 维普 万方数据 等数据库收录!
点击此处可从《浙江大学学报(自然科学版 )》浏览原始摘要信息
点击此处可从《浙江大学学报(自然科学版 )》下载全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司    京ICP备09084417号-23

京公网安备 11010802026262号