首页> 外文会议>Asia Pacific Conference on Postgraduate Research in Microelectronics and Electronics >An efficient VLSI circuit extraction algorithm for transistor-level to gate-level abstraction
【24h】

An efficient VLSI circuit extraction algorithm for transistor-level to gate-level abstraction

机译:一种高效的VLSI电路提取算法,用于晶体管级到门级抽象

获取原文

摘要

This paper proposes an efficient VLSI extraction algorithm to extract a transistor level netlist to a gate level netlist for functional verification and diagnosis. Compared with other reported circuit extraction algorithm, our proposed technique does not require a cell library and is able to generate Boolean equations without the prior knowledge of transistor type or drain/source orientation of the transistors in the netlist. The proposed algorithm firstly transforms the netlist into a matrix. Secondly, the output nodes of different gates are detected using a pre-sorting comparison algorithm. Then, respective pull-up networks (PUNs) and pull-down networks (PDNs) are identified by a depth-first search algorithm. Finally, the function of the PUN/PDN is verified and stored in terms of Boolean equations. The results on 10 standard cells and 6 testing circuits show that the algorithm performs a near-linear time operation. For a 30-transistor standard cell (XOR4), the time required to complete the extraction is 35.4ms, and for a 3639-transistor combinational circuit, the time required to complete the extraction is around 114.4sec. Compared with the reported technique, our proposed algorithm has an average of 8.4% reduction in CPU time.
机译:本文提出了一种高效的VLSI提取算法,将晶体管级网表提取到栅极级网表进行功能验证和诊断。与其他报告的电路提取算法相比,我们所提出的技术不需要小区库,并且能够在没有网上列表中的晶体管类型或晶体管的漏极/源方向的先前知识生成布尔方程。所提出的算法首先将网表转换为矩阵。其次,使用预排序比较算法检测不同门的输出节点。然后,通过深度第一搜索算法识别各个上拉网络(PUNS)和下拉网络(PDNS)。最后,在布尔方程项方面验证了双相/ PDN的功能。 10标准电池和6个测试电路的结果表明该算法执行近线性时间操作。对于30晶体管标准电池(XOR4),完成提取所需的时间为35.4ms,并且对于3639晶体管组合电路,完成提取所需的时间约为114.4分。与报告的技术相比,我们所提出的算法平均降低了CPU时间的8.4%。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号