首页> 外文会议>International Conference on Computer Engineering and Applications >Low Power Binding for Switching Activity Reduction Based on Table Decomposition Method
【24h】

Low Power Binding for Switching Activity Reduction Based on Table Decomposition Method

机译:基于表分解方法的切换活性降低的低功率绑定

获取原文

摘要

Low power design in earlier stage of design process is more effective to reduce dynamic power. In this paper, we focus on the problem of reducing switching activity in high-level synthesis, especially, in binding functional module, bus, register, etc. We propose low power binding algorithm for switching activity reduction based on table decomposition method. The proposed algorithm finds optimal or close-to-optimal binding with fast calculation time by using the property of optimal substructure. Experimental result shows that the proposed method finds a solution 1.7-14.4% more accurately in comparison to conventional heuristic methods, and gets a solution 8.0-479.2 times faster than optimal method does.
机译:设计过程早期阶段的低功耗设计更有效地降低动态功率。在本文中,我们专注于减少高级合成中的切换活动的问题,特别是在绑定功能模块,总线,寄存器等中。我们提出了基于表分解方法的低功率绑定算法进行切换活性降低。该算法通过使用最佳子结构的性质,找到了与快速计算时间的最佳或近距离绑定。实验结果表明,与传统的启发式方法相比,该方法可以更准确地找到1.7-14.4%的解决方案,并比最佳方法更快地获得8.0-479.2倍。

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