CERCIS:一种视频媒体编解码片上系统设计实现

来源 :清华大学学报(自然科学版)网络.预览 | 被引量 : 2次 | 上传用户:chaoschaos
下载到本地 , 更方便阅读
声明 : 本文档内容版权归属内容提供方 , 如果您对本文有版权争议 , 可与客服联系进行内容授权或下架
论文部分内容阅读
基于面向特定应用的可配置处理器架构及其设计方法,设计并完成了一种视频媒体编解码片上系统芯片,它具有通用数字信号处理器的柔性编程及特定目标应用时的高性能等特点。该视频编解码片上系统由编码和解码2部分组成,编码和解码部分都采用相同的媒体信号处理架构。媒体信号处理编码、解码架构中分别包含一个8发射超长指令字数字信号处理器核,还包括实现视频媒体应用的专用数据传输单元,变长编解码单元以及接口单元,可以完成H.263视频媒体编码和解码。在0.13μm工艺库下模拟验证表明,该片上系统在17MHz工作频率下可完成1
其他文献
The eastern Mediterranean region is an active tectonic setting that includes the Dead Sea Transform Fault, which forms the boundary between the African and the Arabian Plates and crosses Lebanon from
期刊
This article summarizes a joint research projec undertaken under the Risk Management Solutions, Inc(RMS) banner to investigate some of the possible approaches for agricultural risk modeling in China.
期刊
The Tohoku Regional Bureau(TRB) of the Ministry of Land, Infrastructure, Transport and Tourism(MLIT) performed various actions in response to the 2011 Great East Japan Earthquake and Tsunami. The tota
期刊
This article addresses three large earthquake disasters in Iran: Tabas in 1978, Rudbar in 1990, and Bam in 2003. Lessons and "Lessons Learned" from these three earthquake disasters were investigated t
期刊
We present preliminary results from the experimental investigation of the response of the atmosphere due to the impact of powerful shock waves.The response is evidenced as ultra low frequency electrom
期刊
Technology innovations are important for firms to compete in the market, whereas spillovers are unavoidable. The research issue of this article is how imitations impact on R&D behaviors. This article
期刊
This paper proposes a kind of facial expression recognition technology based on the D-S theory with weighting coefficient of information fusion, in the background of Affective Computing theory, moreov
期刊
This article presents an efficient and robust algorithm that compute the intersection curve of a ruling surface with a plane. These ruling surface is represented by S(s,t)=B(t)+sV(t) where B(t) is the
期刊
In order to deal with contradiction between big dynamic range and single means to acquiring binarized vertical edge image in texture-based adapting license plate location(TALPL),research on acquiring
期刊
为解决多人行为识别中高维特征空间、角色分配不准确和复杂的时间结构等问题,本文分析了多人行为的特点,提出了一种递归的多层随机网络模型。该模型通过多层网络表达行为的多尺度特性,并由高层体现行为的长时间依赖性。通过对观测的分解大大降低了特征空间的维数,从而降低了问题的复杂度,并在一定程度上消除了目标角色分配不准确带来的影响。实验结果表明本文提出的模型比其他常用模型具有更好的识别效果,即使对复杂行为依然具
期刊