...
首页> 外文期刊>Microelectronics & Reliability >AM3D: An accurate crosstalk probability modeling to predict channel delay in 3D ICs
【24h】

AM3D: An accurate crosstalk probability modeling to predict channel delay in 3D ICs

机译:AM3D:精确的串扰概率模型,可预测3D IC中的通道延迟

获取原文
获取原文并翻译 | 示例

摘要

Migration from Two Dimensional Integrated Circuits (2D ICs) to Three Dimensional Integrated Circuits (3D ICs) reduces the delay due to the shorter wire length between sender and receiver. However, Through-Silicon-Vias (TSVs) that connect layers in the structure of 3D ICs can seriously increase the delay due to capacitance coupling between TSVs and lead to crosstalk fault. The severity of crosstalk faults depends on transitions appearing on TSVs that is called transition patterns. To propose an efficient crosstalk tackling mechanisms in 3D ICs, an accurate probability analytical model is required to predict the delay caused by TSVs (3D ICs) in the attendance of these transition patterns. In this regard, this paper proposes an accurate crosstalk probability model for 3D ICs called AM3D that estimates the latency of communication channels of 3D ICs in the attendance of crosstalk faults. The model is applicable for both non-protected TSVs and TSVs which are protected by crosstalk mitigation mechanisms. In order to accredit our proposed model, various SPICE simulations are conducted and obtained outcomes in different benchmarks are compared with extracted results from AM3D. Comparisons present an average of 1.81% discrepancies between the time delay obtained from SPICE simulations and time delay calculated by our proposed model.
机译:由于发送方和接收方之间的导线长度较短,因此从二维集成电路(2D IC)迁移到三维集成电路(3D IC)可以减少延迟。然而,由于TSV之间的电容耦合,连接3D IC结构中各层的硅通孔(TSV)可能会严重增加延迟,并导致串扰故障。串扰故障的严重程度取决于出现在TSV上的过渡,称为过渡模式。为了在3D IC中提出一种有效的串扰处理机制,需要一种准确的概率分析模型来预测由TSV(3D IC)引起的这些过渡模式引起的延迟。在这方面,本文提出了一种称为AM3D的3D IC准确串扰概率模型,该模型可以估计3D IC通信通道在出现串扰故障时的等待时间。该模型适用于不受保护的TSV和受串扰缓解机制保护的TSV。为了对我们提出的模型进行认证,进行了各种SPICE仿真,并将在不同基准中获得的结果与从AM3D提取的结果进行了比较。比较显示,从SPICE仿真获得的时间延迟与我们提出的模型计算的时间延迟之间平均存在1.81%的差异。

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