| 摘要: |
| 区块链产生的成千上万的交易信息数据不断被收集到区块链中, 会给区块链网络的计算资源和存储空间造成极大负担。为了解决区块链网络计算资源不足和存储空间有限的问题, 我们将边缘计算模式应用于联盟链Hyperledger Fabric系统中,在区块链中设计了基于边缘计算模式的智能合约, 以提高区块链的存储空间和运行效率。在此基础上, 针对区块链存在的数据和用户身份隐私的风险, 提出了基于边缘计算模式的全同态加密智能合约。在执行智能合约过程中, 调用了微软的全同态加密 SEAL 库, 边缘节点的身份信息和区块链的交易数据受到全同态加密算法 BFV 的保护, 在保持交易数据可计算性的同时,实现了交易信息和边缘节点身份的隐私保护。 最后, 经过测试和分析, 边缘计算模式下的原型系统在 14 个边缘节点同时访问时的单节点平均访问时间仅为 139.68 ms, 相比于非边缘计算模式下的原型系统具有更好的性能; 相比目前响应较快的加密联盟链系统 FabZK, 本系统平均用时减少 35.84%, 相比效率提升 55.86%。设计的边缘计算模式下的全同态加密智能合约可以有效提高 Hyperledger Fabric 系统中网络的运行效率, 同时减少了对存储空间的要求, 对于实际环境下的区块链部署, 具有一定参考价值。 |
| 关键词: 边缘计算 智能合约 全同态加密 隐私保护 区块链 |
| DOI：10.19363/J.cnki.cn10-1380/tn.2022.03.10 |
| Received:January 10, 2021Revised:March 25, 2021 |
| 基金项目:本课题得到北京高校高精尖学科建设项目(No.3201023),“十三五”国家密码发展基金(No.MMJJ20170110)资助。 |
|
| Smart Contract with Fully Homomorphic Encryption under Edge Computing Mode |
| YANG Yatao,LIN Tianxiang,CHEN Jianyuan,ZENG Ping,LIU Peihe |
| Department of Electronic and Communication Engineering, Beijing Electronics Science and Technology Institute, Beijing 100070, China;School of Telecommunication Engineering, Xidian University, Xi’an 710071, China |
| Abstract: |
| Thousands of transaction data generated in the blockchain are continuously collected into the blockchain, the computing resources and storage space in blockchain network are becoming great burden. In order to solve the problem of insufficient computing resources and limited storage space in blockchain network, the edge computing model has been introduced to Hyperledger Fabric consortium blockchain system, a smart contract based on edge computing is designed to improve the storage space and operation efficiency in blockchain. After that, in view of the risk of data and user identity privacy in blockchain, a fully homomorphic encryption smart contract based on edge computing mode is proposed. In the process of executing the smart contract, Microsoft's fully homomorphic encryption seal library is adopted. The identity information of edge nodes and transaction data in blockchain are protected by BFV fully homomorphic encryption algorithm. It can not only maintain the transaction data’s computability, but the privacy preservation of transaction data and edge nodes’ identity information also can be guaranteed. Finally, after being tested and analyzed, the average communication time of a single node in the prototype system under the edge computing mode is only 139.68 ms when 14 edge nodes are accessing at the same time, which is much better than the prototype system under the non edge computing mode. Compared with the Fabzk consortium blockchain system with excellent computing performance, the average access time in our prototype system is reduced by 35.84%, and the efficiency is improved by 55.86%. This smart contract with fully homomorphic encryption under edge computing mode can increase the working efficiency of Hyperledger Fabric consortium blockchain system, furthermore, the requirement for storage space will be lower, which provides valuable preferences for blockchain system deployment in real network scene. |
| Key words: edge computing smart contract full homomorphic encryption privacy preservation blockchain |
