面向多用户动态可搜索隐私保护的云存储服务机制

Multiuser dynamic searchable privacy-preserving (MDSPP) is the most reliable security protection for future cloud storage service. This project proposes to investigate and explore the methods, technologies and theory for cloud storage service mechanism on MDSPP. First, a group of attack samples will be established for security threat of clouds search by an experimental verification and a theoretical analysis. With the created attack samples, a security attack model will be constructed from keyword privacy, trapdoor unlinkability privacy, and access patterns privacy. Based on which security attack model, learning from account the secure multi-party computation theory, the security evaluation methods of cloud storage service mechanism on MDSPP will be built. Second, after extracting the general appropriate properties and integrating the most advantages from searchable privacy-preserving mechanism based on symmetric cryptography and public key cryptography which are most suitable for cloud storage service mechanism on MDSPP, a design method will be researched for efficient and flexible search inquiry sentences. Third, by learning from multilevel key-managing mechanism, access control mechanism, and trusted re-encryption secure model, the trusted CSP (cloud service provider) and cloud storage service mechanism on MDSPP will be researched. Finally, an experimental validation platform will be exploited for verifying the creditability and correctness of the proposed cloud storage service mechanism on MDSPP. The validation platform is expected to supply technical and theoretical supports for deploying real system of this cloud storage service mechanism, and theoretical and methodical foundations for further exploring correlated security problem.

多用户动态可搜索隐私保护MDSPP是未来云存储服务最为可靠的安全保障，本项目探讨MDSPP的云存储服务机制设计、评价的方法、技术和相关理论：首先，结合实验验证和理论分析手段，建立针对云搜索的安全威胁样本，从关键词隐私、陷门不可关联性隐私、接入模式隐私抽象出安全威胁模型，借鉴安全多方计算理论思想，探讨MDSPP的云存储服务机制可证明安全评价方法；其次，总结提炼基于对称密码和公钥密码的可搜索隐私保护机制最适合MDSPP的云存储服务机制的特点，融合其优点，研究高效且支持灵活查询语句的可搜索隐私保护的云存储服务机制；第三，借鉴多级密钥管理机制、访问控制机制和可信重加密安全模型，研究可信CSP和MDSPP的云存储服务机制；最后，设计云环境仿真实验平台，验证所设计方案的可信性和正确性，为MDSPP的云存储服务机制的部署提供技术支撑和理论支持，为进一步探讨云存储服务机制的相关安全问题提供基础理论和方案。

多用户动态可搜索隐私保护MDSPP（Multiuser dynamic searchable privacy-preserving）是未来云存储服务最为可靠的安全保障，本项目研究了多用户动态可搜索隐私保护MDSPP的云存储服务机制设计、评价方法、技术和相关理论。重点探讨了安全威胁模型、高效且支持灵活查询语句的可搜索隐私保护的云存储服务机制、支持多用户可搜索隐私保护的云存储服务机制、支持动态可搜索隐私保护的云存储服务机制和仿真实验平台的设计。解决了①如何设计高效灵活的可搜索隐私保护的云存储服务机制？②如何设计不同访问权限的不同等级的密钥管理机制？③如何设计云存储动态更新密文访问控制的可信CSP机制？这3个关键科学问题。主要研究成果发表了5篇SCI期刊论文和4篇EI期刊论文。并将研究成果应用于实际项目开发中，在知识产权方面完成6件软件著作权登记。本研究的成果为MDSPP的云存储服务机制的部署提供了技术支撑和理论支持，为进一步探讨云存储服务机制的相关安全问题提供了基础的理论和方案。