首页 | 本学科首页   官方微博 | 高级检索  
     

基于函数分类和布隆过滤器的布尔匹配方法
引用本文:包杰,王伶俐. 基于函数分类和布隆过滤器的布尔匹配方法[J]. 计算机工程, 2014, 0(6): 275-280
作者姓名:包杰  王伶俐
作者单位:复旦大学专用集成电路与系统国家重点实验室,上海201203
基金项目:国家自然科学基金资助项目(61171011).
摘    要:在现场可编程门阵列的自动化综合流程中,布尔匹配是核心子问题之一。基于布隆过滤器的布尔匹配方法需要消耗大量存储空间并牺牲部分可实现函数的覆盖率。针对该问题,提出一种布尔匹配方法,给出布尔函数的规则表达形式,对布尔函数进行分类,并在布尔匹配的过程中进行动态学习。实验结果表明,通过对函数分类可以使布尔匹配库所需的存储空间降低96%,而动态学习策略可以使电路在逻辑再综合算法应用中额外节省13%的LUT数目。

关 键 词:现场可编程门阵列  布隆过滤器  布尔匹配  函数分类  动态学习  逻辑再综合

Boolean Matching Method Based on Function Classification and Bloom Filter
BAO Jie,WANG Ling-li. Boolean Matching Method Based on Function Classification and Bloom Filter[J]. Computer Engineering, 2014, 0(6): 275-280
Authors:BAO Jie  WANG Ling-li
Affiliation:(State Key Lab ofASIC & Systems, Fudan University, Shanghai 201203, China)
Abstract:In the field of Field Programmable Gate Array(FPGA) automation integrated process, Boolean Matching(BM) is one of the core problems. The Boolean matching method based on Bloom filter needs to consume a large amount of storage space and sacrifices part coverage of realizable function. Aiming at this problem, this paper proposes a Boolean matching method, presents a rule expression form based on Boolean function, which classifies the Boolean function, and carries on dynamic learning in the process of Boolean matching. Experimental results show that, the storage space of Boolean matching can be saved up to 96% by the function classification, and dynamic learning strategy can make the circuit reduce excessively 13% LUT number in application of Logic synthesis algorithm.
Keywords:Field Programmable Gate Array(FPGA)  Bloom filter  Boolean Matching(BM)  function classification  dynamic learning  logic resynthesis
本文献已被 CNKI 维普 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号