Optimization and Design of Eaglesong Hash Based on FPGA

  • Xiao Xiao
  • , Xiaojin Li*
  • , Yabin Sun
  • , Yanling Shi
  • , Tongquan Wei
  • *Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

Based on sponge construction, Eaglesong is a hash algorithm designed for cryptocurrency Nervos CKB. Eaglesong requires a huge number of operations like modular addition, cyclic rotation and Exclusive-OR (XOR) to generate the 256-bit hash data. Considering that the Field-Programmable Gate Array (FPGA) has a great advantage in repetitive operations and parallel data processing, an optimized hardware implementation of Eaglesong on the Xilinx XC7Z100 platform was proposed in this paper. The delay of critical path and frequency were optimized by pipeline design; loop unrolling was used to increase the parallelism and throughput. At the same time, through pre-computation, part of the parameter computation was eliminated, and the number of registers and the occupation of resources were reduced. The results show that the proposed design achieved 238 MHz clock frequency using 191994 slices, and high throughput of 60.93 Gbps.

Original languageEnglish
Title of host publicationProceedings of the 2nd International Conference on Computing and Data Science, CONF-CDS 2021
PublisherAssociation for Computing Machinery
ISBN (Electronic)9781450389570
DOIs
StatePublished - 28 Jan 2021
Event2nd International Conference on Computing and Data Science, CONF-CDS 2021 - Stanford, United States
Duration: 28 Jan 202130 Jan 2021

Publication series

NameACM International Conference Proceeding Series
VolumePartF168982

Conference

Conference2nd International Conference on Computing and Data Science, CONF-CDS 2021
Country/TerritoryUnited States
CityStanford
Period28/01/2130/01/21

Keywords

  • Eaglesong
  • FPGA
  • hash algorithm
  • sponge construction

Fingerprint

Dive into the research topics of 'Optimization and Design of Eaglesong Hash Based on FPGA'. Together they form a unique fingerprint.

Cite this