期刊文献+
任意字段
题名或关键词
题名
关键词
文摘
作者
第一作者
机构
刊名
分类号
参考文献
作者简介
基金资助
栏目信息

嵌入式视频解码器运动补偿的数据布局优化 被引量：1

Data Layout Optimization for Motion Compensation of Embedded Video Decoders
下载PDF
导出
摘要 针对数字信号处理器的片上存取结构特点,提出了一个优化视频解码运动补偿过程数据布局的方案。在片上便签式存储器(SPM)中设置一个乒乓缓冲存储区用于保存运动补偿所需的数据。在对当前宏块进行运动补偿的同时,预取后续运动补偿所需的数据替换其中不再使用的数据;同时,通过数据索引算法获得运动补偿所需的数据的地址,从而实现数据处理与数据存取的并行流水优化。基于TMS320DM642处理器的实验结果表明,优化后MPEG-4视频解码器的解码速度平均提高了6.7%,整个解码过程中DM642的片上二级缓存的能耗平均降低了18.5%。由此可见,对运动补偿过程进行数据布局优化确实可以提高解码性能并降低能耗。 According to the digital signal processor （DSP） chip access structure characteristics, put forward an optimization video decoding motion compensation process data layout scheme. A memory structure of ping-pang buffer is set in Scratch Pad Memory （SPM） for prefetching the data for motion compensation. In the current macro block motion compensation at the same time, prefetch subsequent motion compensation required data replace one no longer use data. In addition, an index algorithm is used to get the address of data for motion compensation, so as to realize data processing and data access parallel flow optimization. The experiments are carded out on a TMS320DM642 platform for a MPEG-4 video decoder. It achieves an increase of about 6.7% in decoding speed and an energy consumption reduction of 18.5% on average for the DM642 level 2 on-chip memory after the optimization. This indicates that the optimization of data layout for motion compensation could enhance the performance and reduce the energy consumption.
作者 郭红星 王恒伟 田婷 张爱华 余胜生
机构地区 华中科技大学计算机科学与技术学院
出处 《计算机技术与发展》 2013年第4期24-28,共5页 Computer Technology and Development
基金 国家自然科学基金资助项目(61173044 60873029) 中央高校基本科研业务费专项资金
关键词 嵌入式系统 流计算 视频解码器 便签式存储器 数据布局 embedded systems stream computing video decoders scratch pad memory data layout
分类号 TP302 [自动化与计算机技术—计算机系统结构]
  • 相关文献

参考文献15

  • 1Panda P R,Catthoor F,Dutt N D,et al. Data and Memory Op- timization Techniques for Embedded Systems[ J]. ACM Trans- actions on Design Automation of Electronic Systems, 2001,6 ( 2 ) : 149-206.
  • 2Kandemir M, Choudhary A. Compiler-directed scratch pad memory hierarchy design and management [ C ]//Proceedings of the 39th Annual Design Automation Conference. NY, USA: ACM ,2002 :628-633.
  • 3黄海林,范东睿,许彤,唐志敏.嵌入式处理器中访存部件的低功耗设计研究[J].计算机学报,2006,29(5):815-821. 被引量：11
  • 4Panda, Ranjan P, Dutt N D, et al. Efficient Utilization of Scratch-pad Memory in Embedded Processor Applications [ C]//Proceedings of the 1997 European Conference on De- sign and Test. Washington, DC, USA : IEEE CS, 1997 : 7-11.
  • 5胡威.基于Scratch Pad Memory的嵌入式系统优化研究[D].杭州:浙江大学,2008.
  • 6Avissar O, Barua R, Stewart D. An optimal memory allocation scheme for scratch-pad-based embedded systems [ J ]. ACM Transactions on Embedded Computer Systems ,2002,1 ( 1 ) :6- 26.
  • 7Cooper K D, Harvey T J. Compiler-controlled memory [ C ]// Proceedings of the eighth international conference on AS- PLOS. NY, USA : ACM, 1998:2-11.
  • 8Ozturk 0 ,Chen G, Kandemir M ,et al. Compiler-directed vari- able latency aware SPM management to cope with timing prob- lems[ C ]//Proe. of CGO "07. Washington, DC, USA: IEEE CS ,2007:232-243.
  • 9Francesco P, Marchalm P, Atienza D, et al. An integrated hardware/software approach for run-time scratchpad manage- ment[Cl//Proc, of41st Annu. DAC. NY,USA:ACM,2004: 238 -243.
  • 10Udayakumaran S, Barua R. Compiler-decided dynamic memo- ry allocation for scratch-pad based embedded systems [ C ]// Proc. of Int. Conf. CASES. NY,USA:ACM,2003:276-286.

二级参考文献12

  • 1Hennessy J.L,Patterson D.A..Computer Architecture:A Quantitative Approach (2nd Edition).Beijing:China Machine Press,2002
  • 2Clark L.T,Choi B,Wilkerson M..Reducing translation lookaside buffer active power.In:Proceedings of the ISLPED,Seoul,Korea,2003,10～13
  • 3Juan T.et al.Reducing TLB power requirements.In:Proceedings of the ISLPED,Monterey,California,1997,196～201
  • 4Chen J.B.et al.A simulation based study of TLB performance.In:Proceedings of the ISCA,Queensland,Australia,1992,114～123
  • 5Inoue K..High-performance low-power cache memory architecture[Ph.D.dissertation].Kyushu University,Fukuoka,Japan,2001
  • 6Chang Yen-Jen,Ruan Shang-Jang,Lai Fei-Pei.Design and analysis of low-power cache using two-level filter scheme.IEEE Transactions on VLSI Systems,2003,11(4):568～580
  • 7Kin J,Gupta M,Mangione-Smith W.H..The filter cache:An energy efficient memory structure.In:Proceedings of the MICRO-97:ACM/IEEE International Symposium on Microarchitecture,Los Alamitos,USA,1997,184～193
  • 8Zhu Zhi-Chun,Zhang Xiao-Dong.Access-mode predictions for low-power cache design.IEEE Micro,2002,22(2):58～71
  • 9Park Gi-Ho,Lee Kil-Whan et al.A low-power cache system for embedded processors.In:Proceedings of the 43th IEEE Midwest Symposium on Circuits and Systems,Monterey,Canada,2000,316～319
  • 10Gowan M.K,Biro L.L,Jachson D.B..Power considerations in the design of the alpha 21264 microprocessor.In:Proceedings of the 35th Annual Design Automation Conference (DAC'98),Los Alamitos,USA,1998,26～31

共引文献10

同被引文献14

  • 1Steven T.Graves.如何实现对MP3播放器嵌入式软件中列表数据的管理?[J].中国科技信息,2006(7):71-72. 被引量：1
  • 2File allocation table [ EB/OL ]. 2013. http ://en. wikipedia. org/wiki/File Allocation_Table.
  • 3Microsoft extensible firmware initiative FAT32 file system sp- ecification FAT: General overview of on-disk format, hard- ware white paper,version 1.03[ M]. [s. l. ]: [s. n. ] ,2000.
  • 4Aucouturier J J, Pachet F. Scaling up music playlist generation [ C]//Proc of IEEE international conference on multimedia expo. [ s. l. ]: [ s. n. ] ,2002.
  • 5Logan B. Content-based playlist generation:Exploratory ex- periments[ C]//Proc of ISMIR. [s. l. ] :Is. n. ] ,2002.
  • 6Liu Hao, Hu Jun, Rauterberg M. Music playlist recommenda- tion based on user heartbeat and music preference[ C]//Proc of international conference on computer technology and devel- opment. [ s. l. ] : [ s. n. ] ,2009.
  • 7Liu Hao, Hu Jun, Rauterberg M. Software architecture support for biofeedback based in-flight music systems [ C ]//Proc of 2nd IEEE international conference on computer science and information technology. [ s. l. ]:[ s. n. ] ,2009.
  • 8Chiarandini L, Zanoni M, Sarti A. A system for dynamic playl- ist generation driven by multimodal control signals and de-scriptors[ C]//Proc of IEEE international workshop on multi- media signal processing. [s. 1. ] : [s. n. ] ,2011.
  • 9Pampalkl E,Pohlel T, Widmer G. Dynamic playlist generation based on skipping behavior[ C ]//Proc of the 6th ISMIR con- ference. [s. 1. ] :[s. n. ] ,2005.
  • 10van Gulik R, Vignoli F. Visual playlist generation on the artist map[ C]//Proc of international conf on music information re-trieval. London,UK: [ s. n. ] ,2005.

引证文献1

二级引证文献1

计算机技术与发展
2013年 第4期

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部