{"title":"Design of Bitstream Co-processor for Multimedia Applications","authors":"Yingke Gao, Huan Ying, Zhiyuan Xue, Tiejun Zhang, Donghui Wang, C. Hou","doi":"10.1109/DASC.2013.66","DOIUrl":null,"url":null,"abstract":"Variable length coding (VLC) is widely used in the entropy codec part of various video codec standards. Its parallelism is difficult to be developed due to its inherent bit-serial characteristic. As it becomes the system bottleneck of real-time video decoding, this paper presents a bitstream co-processor which greatly accelerates the codec of VLC by designing corresponding bitstream instructions. With SuperV DSP as the research prototype, the area cost of the bitstream co-processor only increases by 1.71% while the performance of the application of entropy decoding is improved about 73% on average.","PeriodicalId":179557,"journal":{"name":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","volume":"35 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2013-12-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"2013 IEEE 11th International Conference on Dependable, Autonomic and Secure Computing","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/DASC.2013.66","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Variable length coding (VLC) is widely used in the entropy codec part of various video codec standards. Its parallelism is difficult to be developed due to its inherent bit-serial characteristic. As it becomes the system bottleneck of real-time video decoding, this paper presents a bitstream co-processor which greatly accelerates the codec of VLC by designing corresponding bitstream instructions. With SuperV DSP as the research prototype, the area cost of the bitstream co-processor only increases by 1.71% while the performance of the application of entropy decoding is improved about 73% on average.