首页> 外文期刊>電子情報通信学会技術研究報告. 集積回路. Integrated Circuits and Devices >New Architecture for Multiple-Valued Fine-Grain Reconfigurable VLSI Based on Current-Mode Logic
【24h】

New Architecture for Multiple-Valued Fine-Grain Reconfigurable VLSI Based on Current-Mode Logic

机译:基于电流模式逻辑的多值细粒度可重构VLSI的新架构

获取原文
获取原文并翻译 | 示例
           

摘要

This article presents a fine-grain reconfigurable VLSI based on multiple-valued X-net data transfer scheme. Two binary data can be transferred from two cells to one common adjacent cell simultaneously at each X intersection, which leads to area-efficient data transfer without reducing performance. Each cell is composed of a switch block and a logic block. Multiple-valued signaling is utilized to implement a compact switch block. In each logic block, differential-pair circuits are used to realize low-power logic operations including threshold operations, an arbitrary 2-variable binary function and a bit-serial addition. Moreover, a current-source sharing technique between a series-gating differential-pair circuit and a current-mode D-latch is utilized to reduce the current source count for low power. As a result, the area and power consumption of the multiple-valued reconfigurable VLSI is reduced to 60% and 82% in comparison with that of an equivalent CMOS reconfigurable VLSI, respectively, while the delay is kept same.
机译:本文提出了一种基于多值X-net数据传输方案的细粒度可重新配置的VLSI。可以在每个X交叉点同时将两个二进制数据从两个单元传输到一个公共的相邻单元,从而在不降低性能的情况下实现了区域有效的数据传输。每个单元由一个开关模块和一个逻辑模块组成。利用多值信令来实现紧凑的开关块。在每个逻辑块中,差分对电路用于实现低功耗逻辑运算,包括阈值运算,任意2变量二进制函数和位串行加法。此外,利用串联选通差分对电路和电流模式D锁存器之间的电流源共享技术来减少低功耗的电流源数量。结果,与相同的CMOS可重构VLSI相比,多值可重构VLSI的面积和功耗分别减小到60%和82%,同时延迟保持不变。

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