首页> 外文会议>IEEE Pulsed Power Conference >Sub-ps Resolution Programmable Delays Implemented in a Xilinx FPGA
【24h】

Sub-ps Resolution Programmable Delays Implemented in a Xilinx FPGA

机译:在Xilinx FPGA中实现的子PS分辨率可编程延迟

获取原文

摘要

In this paper, a novel way to finely tune a net delay on Xilinx Field Programmable Gate arrays (FPGAs) is proposed. It consists of adding floating interconnects (nodes) to the net on which the delay is to be tuned, connected to any input pin of a switch matrix along the net. Adding nodes is made with a TCL script applied to an already placed and routed design. However, such nodes, also called antennas, typically cause fatal errors during the design flow and normally prevent the tools from generating the bit stream. To overcome this issue, a breadth-first search algorithm connecting each node to a load is proposed in this work. Experimental results conducted on a ZYNQxc7z010-3clg400 Xilinx FPGA using the Vivado Design suite showed that it is possible to add small delay steps to a net with a resolution under a pico-second and a covered range proportional to the number of added nodes reaching 48.6 ps for a net with 15 added nodes.
机译:本文提出了一种精细调整Xilinx场可编程门阵列(FPGA)上的净延迟的新方法。它包括将浮动互连(节点)添加到要调谐的延迟的网络上,连接到沿网的开关矩阵的任何输入引脚。添加节点是使用应用于已经放置和路由设计的TCL脚本进行的。然而,这种节点也称为天线,通常在设计流程期间导致致命的误差,并且通常防止工具产生比特流。为了克服这个问题,在这项工作中提出了一种将每个节点连接到负载的广度第一搜索算法。使用Vivado设计套件在ZynqXC7Z010-3CLG400 Xilinx FPGA上进行的实验结果表明,可以在微微第二和覆盖范围下与达到48.6 ps的增加节点的数量成比例的覆盖范围添加小延迟步骤。达到48.6 ps的数量对于带有15个添加节点的网络。

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