电子学报 ›› 2022, Vol. 50 ›› Issue (6): 1399-1409.DOI: 10.12263/DZXB.20200731

• 学术论文 • 上一篇    下一篇

面向CNN的区块链可信隐私服务计算模型

丁毅1, 沈薇1, 李海生2,3, 钟琼慧1, 田明宇1, 李洁1   

  1. 1.北京物资学院信息学院,北京 101149
    2.北京工商大学农产品质量安全追溯技术及应用国家工程实验室,北京 100048
    3.北京工商大学计算机学院,北京 100048
  • 收稿日期:2020-07-16 修回日期:2021-03-30 出版日期:2022-06-25
    • 通讯作者:
    • 李洁
    • 作者简介:
    • 丁 毅 男,1981年9月出生于河北省沧州市.北京物资学院信息学院副教授,硕士生导师.主要研究方向为区块链和智能合约技术、隐私计算等.
      沈 薇 女,1999年9月出生于江苏省宿迁市.北京物资学院本科生.主要研究方向为同态加密和区块链技术.
      沈 薇 女,1999年9月出生于江苏省宿迁市 . 北京物资学院本科生 . 主要研究方向为同态加密和区块链技术.
      李 洁(通讯作者) 女,1983年1月出生于北京市.北京物资学院信息学院助理研究员.主要研究方向为区块链、隐私计算及形式化验证等.
    • 基金资助:
    • 国家重点研发计划 (2018YFB1402703); 北京市教育委员会科技计划一般项目 (KM201910037003); 北京工商大学农产品质量安全追溯技术及应用国家工程实验室开放课题 (AQT-2020-YB5); 北京市社会科学基金研究基地项目 (18JDGLB026); 北京物资学院2020年度“实培计划”项目

Blockchain Trusted Privacy Service Computing Model for CNN

DING Yi1, SHEN Wei1, LI Hai-sheng2,3, ZHONG Qiong-hui1, TIAN Ming-yu1, LI Jie1   

  1. 1.School of Information, Beijing Wuzi University, Beijing 101149, China
    2.National Engineering Laboratory for Agri-product Quality Traceability, Beijing Technology and Business University, Beijing 100048, China
    3.School of Computer Science and Engineering, Beijing Technology and Business University, Beijing 100048, China
  • Received:2020-07-16 Revised:2021-03-30 Online:2022-06-25 Published:2022-06-25
    • Corresponding author:
    • LI Jie
    • Supported by:
    • National Key Research and Development Program of China (2018YFB1402703); General Program of Science and Technology Planning Project of Beijing Municipality Education Commission (KM201910037003); Open Project of State Engineering Laboratory of Agricultural Product Quality and Safety Traceability Technology and Application, Beijing Technology and Business University (AQT-2020-YB5); Research Base Project of Beijing Municipal Social Science Foundation (18JDGLB026); Practical Training Project of Beijing Wuzi University in 2020

摘要:

在当前移动互联网时代,数据量增长迅速,服务计算能力不断增强,数据隐私保护和服务环境可信成为备受关注的重要问题.本文研究面向卷积神经网络典型应用场景的可信隐私服务计算模型,探索支持同态加密的数据和模型计算方法,保护数据隐私.构建基于区块链和智能合约技术服务过程存证及计算权益分配方法,保证服务计算的公开透明、可信可追溯.探索资源提供者、模型拥有者及用户的新型云环境资源数据服务模式,促进资源有效整合,发展共享经济.最后,通过实验分析该模型的隐私保护方法.

关键词: 同态加密, 隐私保护, 智能合约, 卷积神经网络

Abstract:

Data privacy protection and service environment trust have become important issues with the rapidly increasing amount of data and computing power of services in current era of Mobile Internet. This paper studies the trusted privacy service computing model for typical application scenarios of convolutional neural network. It explores data and model computing methods supporting homomorphic encryption to protect data privacy, builds methods of service certificate storage and calculating equity interests distribution based on blockchain and smart contract technology to ensure the openness, transparency, credibility and traceability of service computing. A novel resource and data service paradigm of cloud environment is explored for resource providers, model owners and users to promote the effective integration of resources and develop sharing economy. Finally, the privacy protection method in the model is analyzed through experiments.

Key words: homomorphic encryption, privacy protection, smart contract, convolutional neural network

中图分类号: