边缘区块链场景ML-DSA 数字签名并行优化研究

刘建美 ,  王娅如

信息科学与工程研究 ›› 2026, Vol. 7 ›› Issue (6) : 99 -102.

PDF (1184KB)
信息科学与工程研究 ›› 2026, Vol. 7 ›› Issue (6) : 99 -102. DOI: 10.12349/iser.v7i6.10813

边缘区块链场景ML-DSA 数字签名并行优化研究

作者信息 +

Parallel Optimization of ML-DSA Digital Signatures for Edge Blockchain Scenarios

Author information +
文章历史 +
PDF (1212K)

摘要

量子计算对传统公钥密码体系形成安全威胁,NIST标准化的ML-DSA是主流后量子签名算法,但签名计算开销大,难以部署在算力受限的边缘区块链节点.本文搭建 Python-PQCrypto 混合架构,提出基于进程池的动态任务分片并行优化方案,借助多进程机制规避Python全局解释器锁带来的单核执行限制,实现多核批量签名并行运算.在16核Windows平台开展测试,优化后的ML-DSA-44签名吞吐量由1225.82 TPS提升至6738.42 TPS,加速比达5.50倍.区块链仿真场景下,5000笔交易负载时系统吞吐量稳定1937.76 TPS,平均签名时延0.430 ms、验签时延0.085 ms,在无传输干扰的测试条件下签名与验签校验成功率达100%.该优化方案有效缓解格基签名算力瓶颈,无需专用硬件加速,可直接适配现有多核边缘设备.

Abstract

The quantum computing threat renders traditional public-key cryptography insecure, and ML-DSA standardized by NIST becomes the core post-quantum signature scheme. However, its high signing overhead restricts deployment on resource-limited edge blockchain nodes. This paper constructs a Python-PQCrypto hybrid architecture, proposes a dynamic task partitioning parallel optimization method based on multi-process pool to bypass Python GIL and realize true multi-core parallel batch signature computation. Experiments on a 16-core Windows platform show that the throughput of optimized ML-DSA-44 rises from 1225.82 TPS to 6738.42 TPS with a speedup of 5.50×. In blockchain simulation, the system maintains 1937.76 TPS under 5000 transactions, with average signing latency 0.430 ms and verification latency 0.085 ms, and 100% signature success rate. The software parallel optimization scheme effectively relieves the lattice signature computing bottleneck, which can be deployed on existing multi-core edge devices without dedicated cryptographic accelerators.

关键词

边缘区块链 / ML-DSA / 后量子密码 / 并行优化

Key words

edge blockchain / ML-DSA / post-quantum cryptography / parallel optimization

引用本文

引用格式 ▾
刘建美,王娅如. 边缘区块链场景ML-DSA 数字签名并行优化研究[J]. 信息科学与工程研究, 2026, 7(6): 99-102 DOI:10.12349/iser.v7i6.10813

登录浏览全文

4963

注册一个新账户 忘记密码

参考文献

[1]

陈驰, 马建峰. 后量子密码标准化进展与区块链适配关键问题[J]. 通信学报, 2024, 45(04): 132-148.

[2]

National Institute of Standards and Technology. FIPS 204: Module-Lattice-Based Digital Signature Standard(ML-DSA)[S]. Gaithersburg: NIST, 2024. https://doi.org/10.6028/NIST.FIPS.204

[3]

刘建伟, 张宗洋. 边缘区块链安全体系架构与格基密码部署挑战[J]. 计算机学报, 2024, 47(07): 1561-1580.

[4]

Longa P, Naehrig M. Speeding up the Number Theoretic Transform for Faster Ideal Lattice-Based Cryptography[J]. Designs, Codes and Cryptography, 2022, 90(8): 1741-1770. https://doi.org/10.1007/s10623-022-01089-9

[5]

杨波, 王瑞锦. 边缘计算场景下区块链交易并行处理优化方法[J]. 信息科学与工程研究, 2024, 11(02): 38-46.

基金资助

河南省科技攻关计划项目(242102210189)

河南省自然科学基金青年项目(252300423360)

河南省高等学校重点科研项目(26A520052)

中原工学院科研团队发展项目(K2025TD002)

国家留学基金项目(202508410108)

AI Summary AI Mindmap
PDF (1184KB)

7

访问

0

被引

详细

导航
相关文章

AI思维导图

/

〈 〉