智能物联节点云连接中匿名认证及全同态加密技术研究

As well as the computation capability and storage capacity of smart Internet of Things (IoT) node are enhanced, new bidirectional anonymous authentication protocol and symmetrical full homomorphic encryption algorithm are proposed. Based on trusted computing and full homomorphic encryption theories, the proposed secure technologies can guarantee the security of those unattended and unmanned smart IoT platforms and their data transmission when they connect to cloud computing center. The identities of trusted root and host are bundled authenticated by Issuer and Verifier, while the smart IoT node will authenticate the Issuer and Verifier too. Combined with those protocol simplify measures, the proposed scheme will effectively solve the secure problems when smart IoT node is stealed, cheated or attacked. Symmetrical full homomorphic encryption algorithm is proposed to realize data security and service privacy when smart IoT node connects to cloud center. The recryption mechanism, gradually expansion of evaluate function and key negotiation protocol are improved in the proposed full homomorphic encryption algorithm. Based on Random Oracle Model, security proofs of the proposed bidirectional anonymous authentication protocol and symmetrical full homomorphic encryption key negotiation protocol are done, while the efficiency tests of the proposed protocols and algorithms are carried out on computers and smart IoT nodes. The obtained research results of this project are expected to effectively solve the platform authentication and data security when machines are connected to cloud center, and help to construct secure smart IoT connection system in the future.

针对未来智能物联节点运算存储能力提高以及云接入需求迫切的特点，在可信计算及全同态加密理论基础上，提出了新的双向匿名认证和对称全同态加密的安全技术来保证无人值守、无人操作的智能物联节点连入云计算中心过程中的平台及传输安全。通过增加可信根与主机的ID捆绑认证，节点对证书发行方及平台认证方的反向认证，结合更多协议化简措施，有效解决智能物联节点被盗取、欺骗、攻击等安全问题。研究对称全同态加密算法来实现智能物联节点接入云端系统过程中的数据安全及隐私服务问题。改进重加密机制、逐步拓展允许函数范围，并为对称全同态加密提出适合智能物联系统的密钥协商机制。在随机预言机模型下对所提双向匿名认证协议及全同态加密密钥协商协议进行安全证明，在计算机和智能物联节点上对提出的协议及算法进行效率测试实验。本课题预计获得的研究成果有效解决机器到云连接过程中的平台认证及数据安全，有助于未来智能物联系统的安全搭建。

针对未来智能物联节点运算存储能力提高以及云接入需求迫切的特点，在可信计算及全同态加密理论基础上，提出了新的双向匿名认证和全同态加密的安全技术来保证无人值守、无人操作的智能物联节点连入云计算中心过程中的平台及传输安全。针对预定的研究目标，在平台认证方面，我们提出了适用于智能物联网云接入的MA-DAA远程匿名认证协议，首次提出有效方案解决这一方面的安全问题。另外我们基于UC安全框架提出改进的UC-DAA平台远程匿名认证协议，在安全性和可用性方面进行了提升，同时建立了智能物联节点接入云的安全框架演示系统进行仿真和验证。在全同态算法研究方面，提出了QPC-PKC SWHE方案来有效压缩全同态加密过程中的密钥存储空间，另外提出了PKCB-FHE方案来进一步提高全同态加密效率。另外，在安全数据融合、全同态加密深度优化、LWE全同态加密、远程匿名认证、门限群签名等方面提出了有效方法和改进方案。取得的研究成果在IEEE Access、International Journal of Theoretical Physics、AEU-International Journal of Electronics and Communications、INSCRYPT 2017、GLOBECOM 2016、ICT 2014、IIKI 2014、WCSN 2016、Journal of Southeast University、计算机学报、东南大学学报、南京邮电大学学报等国际国内期刊和学术会议上发表了12篇论文，另外还有3篇论文已录用待发表；申请发明专利7项，其中已获得授权2项，提交1项国际标准提案；培养了12名研究生，其中已毕业11人。本课题预计获得的研究成果有效解决机器到云连接过程中的平台认证及数据安全，有助于未来智能物联系统的安全搭建。