[1]刘波,季程,朱婉瑜,等.基于配置信息动态压缩的可重构架构优化[J].东南大学学报(自然科学版),2015,45(5):822-827.[doi:10.3969/j.issn.1001-0505.2015.05.002]
　Liu Bo,Ji Cheng,Zhu Wanyu,et al.Optimization of reconfigurable processor based on dynamic configuration context compression[J].Journal of Southeast University (Natural Science Edition),2015,45(5):822-827.[doi:10.3969/j.issn.1001-0505.2015.05.002]
点击复制

基于配置信息动态压缩的可重构架构优化()
分享到：

《东南大学学报(自然科学版)》[ISSN:1001-0505/CN:32-1178/N]

卷:
45
期数:
2015年第5期
页码:
822-827
栏目:
电子科学与工程
出版日期:
2015-09-20

文章信息/Info

Title:
Optimization of reconfigurable processor based on dynamic configuration context compression
作者:
刘波; 季程; 朱婉瑜; 梅晨
东南大学国家专用集成电路系统工程技术研究中心, 南京 210096
Author(s):
Liu Bo;  Ji Cheng;  Zhu Wanyu;  Mei Chen
National ASIC System Engineering Technology Research Center, Southeast University, Nanjing 210096, China
关键词:
雷达应用;  粗粒度可重构处理器;  配置信息压缩;  重构性能
Keywords:
radar application;  coarse grained reconfigurable processor;  configuration compression;  reconfiguration performance
分类号:
TN302
DOI:
10.3969/j.issn.1001-0505.2015.05.002
摘要:
提出了一种基于动态模式匹配的配置信息压缩方法及其硬件实现方案,通过去除配置信息中的冗余,提升了粗粒度可重构处理器的重构性能.基于SMIC 40 nm工艺,完成了所述优化方案的电路实现与验证,系统工作主频为500 MHz.针对雷达应用核心算子的实验结果表明,所提的可重构架构优化方法使雷达应用算子配置信息量缩减了60%以上,相比SIMD压缩方法高出近50%.此外,所提方法的解压缩时间仅需6 ns,其解压缩性能优于SIMD压缩方法和字典压缩方法.
Abstract:
A configuration compression approach based on dynamic pattern matching and its hardware implementation are proposed to efficiently remove the redundancies in configuration contexts and thus improve the reconfiguration performance of coarse grained reconfigurable processors. The experimentation of the proposed approach is implemented based on a SMIC(semiconductor manufacturing international corporation)40 nm process. The system working frequency is 500 MHz. The experimental results of the case kernel algorithms for radar application show that the proposed compression approach can reduce the context size by over 60%, which is about 50% higher than that of the SIMD(single instruction multiple data)approach. Besides, the de-compression time of the proposed approach is only 6 ns, and the de-compression performance is better than those of the SIMD approach and the dictionary approach.

参考文献/References:

[1] Anjum O, Ahonen T, Garzia F, et al. State of the art baseband DSP platforms for software defined radio: a survey[J]. EURASIP Journal on Wireless Communications and Networking, 2011, 2011(5):5-24.
[2] Ganesan M K A, Singh S, May F, et al. H.264 decoder at HD resolution on a coarse grain dynamically reconfigurable architecture[C]//2007 International Conference on Field Programmable Logic and Applications. London, UK, 2007:467-471.
[3] Ahmadinia A. Integrated scheduling and configuration caching in dynamically reconfigurable systems[C]//2010 International Conference on Intelligent and Advanced Systems. Kuala Lumpur, Malaysia, 2010:1-6.
[4] Liu B, Zhu W Y, Liu Y, et al. A configuration compression approach for coarse-grain reconfigurable architecture for radar signal processing[C]//2014 International Conference on Cyber-Enabled Distributed Computing and Knowledge Discovery. Shanghai, China,2014:448-453.
[5] Wang Y, Liu L, Yin S, et al. On-chip memory hierarchy in one coarse-grained reconfigurable architecture to compress memory space and to reduce reconfiguration time and data-reference time[J].IEEE Transactions on Very Large Scale Integration Systems, 2014, 22(5):983-994.
[6] Thoma F, Kühnle M, Bonnot P, et al. MORPHEUS: heterogeneous reconfigurable computing[C]//2007 International Conference on Field Programmable Logic and Applications. Amsterdam, Netherlands, 2007:409-414.
[7] Kim Y, Mahapatra R N. Dynamic context compression for low-power coarse-grained reconfigurable architecture[J]. IEEE Transactions on Very Large Scale Integration Systems, 2010, 18(1):15-28.
[8] Park S, Choi K. An approach to code compression for CGRA[C]//2011 3rd Asia Symposium on Quality Electronic Design. Kuala Lumpur, Malaysia, 2011:240-245.
[9] Jafri S M A H, Hemani A, Paul K, et al. Compression based efficient and agile configuration mechanism for coarse grained reconfigurable architectures[C]//2011 IEEE International Symposium on Parallel and Distributed Processing Workshops and PHD Forum. Anchorage, Alaska, USA, 2011:290-293.
[10] Murashita K, Satoh N, Okada Y, et al. High-speed statistical compression using self-organized rules and predetermined code tables[C]//Proceedings of 1996 IEEE Data Compression Conference. Snowbird, UT, USA, 1996:449.

备注/Memo

备注/Memo:
收稿日期: 2015-03-29.
作者简介: 刘波(1984—),男,博士,讲师,liubo_cnasic@seu.edu.cn.
基金项目: 国家自然科学基金资助项目(61204023,61404028)、国家高技术研究发展计划(863计划)资助项目(2012AA012703).
引用本文: 刘波,季程,朱婉瑜,等.基于配置信息动态压缩的可重构架构优化[J].东南大学学报:自然科学版,2015,45(5):822-827. [doi:10.3969/j.issn.1001-0505.2015.05.002]
更新日期/Last Update: 2015-09-20