首页> 外文会议>2016 IEEE Annual India Conference >Design of high speed CRC algorithm for ethernet on FPGA using reduced lookup table algorithm
【24h】

Design of high speed CRC algorithm for ethernet on FPGA using reduced lookup table algorithm

机译:利用简化查找表算法设计FPGA上以太网高速CRC算法

获取原文
获取原文并翻译 | 示例

摘要

This paper describes the design and development of modified CRC algorithm for the hardware implementation on FPGA to meet the speed constraint for Ethernet, using the reduced lookup table algorithm. This algorithm can be applied for any length of data, by processing it in a block of 16 bytes at a time. The last block may have less than 16 bytes. To process an input block of 16 bytes, the algorithm first forms an optimized table of pre-calculated CRC. Corresponding to the input data, lookup from this table is done and the results from the table lookup are combined by XOR operations to form the final CRC of the input data. The Ethernet data whose CRC needs to be calculated is processed in blocks of 128 bits at clock frequency of 312.5 MHz to achieve a throughput of 40Gbps. The entire design is functionally verified using ModelSim SE Plus 6.3g. Applications in Internet of Things and Machine-to-Machine require real time big data platforms and Artificial Intelligence platforms, where there is high demand for lower latency and high speed network infrastructure. To create such a reliable network infrastructure at high speeds, a hardware accelerated CRC error detection needs to be used.
机译:本文介绍了使用简化的查找表算法,针对FPGA上的硬件实现以满足以太网速度限制的改进CRC算法的设计和开发。通过一次以16字节的块进行处理,该算法可以应用于任何长度的数据。最后一块可能少于16个字节。为了处理一个16字节的输入块,该算法首先形成一个预先计算的CRC的优化表。对应于输入数据,完成从该表中的查找,并且通过XOR操作将来自表查找的结果组合起来,以形成输入数据的最终CRC。需要计算CRC的以太网数据以312.5 MHz的时钟频率以128位的块进行处理,以实现40Gbps的吞吐量。整个设计均使用ModelSim SE Plus 6.3g进行了功能验证。物联网和机器对机器中的应用程序需要实时大数据平台和人工智能平台,在这些平台上,对低延迟和高速网络基础架构的需求很高。为了高速创建这种可靠的网络基础结构,需要使用硬件加速的CRC错误检测。

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