Compressor tree based processing element optimization in propagate partial SAD architecture

Yiqing Huang, Qin Liu, T. Ikenaga
{"title":"Compressor tree based processing element optimization in propagate partial SAD architecture","authors":"Yiqing Huang, Qin Liu, T. Ikenaga","doi":"10.1109/APCCAS.2008.4746388","DOIUrl":null,"url":null,"abstract":"In H.264/AVC standard, the improvement of motion estimation (ME) part helps to enhance the performance greatly. However, the ME part, especially the integer motion estimation (IME) occupies computation complexity dramatically, which leads to complexity in hardware implementation. Many works have been done to achieve efficient IME engine and propagate partial SAD (PPSAD) architecture is the most efficient one in data path and hardware cost. Based on PPSAD structure, this paper proposes a compressor tree based compact PE array architecture. The 4-2 and 3-2 compressor trees are used to build up this compact structure. The proposed structure is embedded into PPSAD architecture and synthesized under different frequency points. With TSMC 0.18 mum 1P8M technology, the proposed architecture can achieve 10%-13% hardware cost reduction for a single 4times4 PE array compared with most recent work. About 10.7 k, 13.2 k and 6.5 k gates hardware cost can be saved compared with previous PPSAD structures.","PeriodicalId":344917,"journal":{"name":"APCCAS 2008 - 2008 IEEE Asia Pacific Conference on Circuits and Systems","volume":"38 1","pages":"0"},"PeriodicalIF":0.0000,"publicationDate":"2008-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"8","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"APCCAS 2008 - 2008 IEEE Asia Pacific Conference on Circuits and Systems","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.1109/APCCAS.2008.4746388","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 8

Abstract

In H.264/AVC standard, the improvement of motion estimation (ME) part helps to enhance the performance greatly. However, the ME part, especially the integer motion estimation (IME) occupies computation complexity dramatically, which leads to complexity in hardware implementation. Many works have been done to achieve efficient IME engine and propagate partial SAD (PPSAD) architecture is the most efficient one in data path and hardware cost. Based on PPSAD structure, this paper proposes a compressor tree based compact PE array architecture. The 4-2 and 3-2 compressor trees are used to build up this compact structure. The proposed structure is embedded into PPSAD architecture and synthesized under different frequency points. With TSMC 0.18 mum 1P8M technology, the proposed architecture can achieve 10%-13% hardware cost reduction for a single 4times4 PE array compared with most recent work. About 10.7 k, 13.2 k and 6.5 k gates hardware cost can be saved compared with previous PPSAD structures.
基于压缩器树的传播部分SAD结构处理单元优化
在H.264/AVC标准中,对运动估计(ME)部分的改进有助于大幅度提高性能。然而,运动估计部分,尤其是整数运动估计(IME),极大地占用了计算量,从而导致硬件实现的复杂性。为了实现高效的IME引擎,已经做了大量的工作,并且在数据路径和硬件成本方面,PPSAD (partial SAD)架构是最有效的。在PPSAD结构的基础上,提出了一种基于压缩树的紧凑PE阵列结构。4-2和3-2压缩树用于构建这种紧凑的结构。该结构被嵌入到PPSAD架构中,并在不同的频率点下进行合成。采用台积电0.18 mum 1P8M技术,与最近的工作相比,该架构可以将单个4times4 PE阵列的硬件成本降低10%-13%。与以前的PPSAD结构相比,可节省约10.7 k, 13.2 k和6.5 k的栅极硬件成本。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 求助全文
来源期刊
自引率
0.00%
发文量
0
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信