With the development of cloud computing, the mutual understandability among distributed data access control has become an important issue in the security field of cloud computing. To ensure security, confidentiality a...With the development of cloud computing, the mutual understandability among distributed data access control has become an important issue in the security field of cloud computing. To ensure security, confidentiality and fine-grained data access control of Cloud Data Storage (CDS) environment, we proposed Multi-Agent System (MAS) architecture. This architecture consists of two agents: Cloud Service Provider Agent (CSPA) and Cloud Data Confidentiality Agent (CDConA). CSPA provides a graphical interface to the cloud user that facilitates the access to the services offered by the system. CDConA provides each cloud user by definition and enforcement expressive and flexible access structure as a logic formula over cloud data file attributes. This new access control is named as Formula-Based Cloud Data Access Control (FCDAC). Our proposed FCDAC based on MAS architecture consists of four layers: interface layer, existing access control layer, proposed FCDAC layer and CDS layer as well as four types of entities of Cloud Service Provider (CSP), cloud users, knowledge base and confidentiality policy roles. FCDAC, it’s an access policy determined by our MAS architecture, not by the CSPs. A prototype of our proposed FCDAC scheme is implemented using the Java Agent Development Framework Security (JADE-S). Our results in the practical scenario defined formally in this paper, show the Round Trip Time (RTT) for an agent to travel in our system and measured by the times required for an agent to travel around different number of cloud users before and after implementing FCDAC.展开更多
Big data has a strong demand for a network infrastructure with the capability to support data sharing and retrieval efficiently. Information-centric networking （ICN） is an emerging approach to satisfy this demand, w...Big data has a strong demand for a network infrastructure with the capability to support data sharing and retrieval efficiently. Information-centric networking （ICN） is an emerging approach to satisfy this demand, where big data is cached ubiquitously in the network and retrieved using data names. However, existing authentication and authorization schemes rely mostly on centralized servers to provide certification and mediation services for data retrieval. This causes considerable traffic overhead for the secure distributed sharing of data. To solve this problem, we employ identity-based cryptography （IBC） to propose a Distributed Authentication and Authorization Scheme （DAAS）, where an identity-based signature （IBS） is used to achieve distributed verifications of the identities of publishers and users. Moreover, Ciphertext-Policy Attribnte-based encryption （CP-ABE） is used to enable the distributed and fine-grained authorization. DAAS consists of three phases： initialization, secure data publication, and secure data retrieval, which seamlessly integrate authentication and authorization with the in- terest/data communication paradigm in ICN. In particular, we propose trustworthy registration and Network Operator and Authority Manifest （NOAM） dissemination to provide initial secure registration and enable efficient authentication for global data retrieval. Meanwhile, Attribute Manifest （AM） distribution coupled with automatic attribute update is proposed to reduce the cost of attribute retrieval. We examine the performance of the proposed DAAS, which shows that it can achieve a lower bandwidth cost than existing schemes.展开更多
为解决数据混合存储导致精准查找速度慢、数据未分类分级管理造成安全治理难等问题,构建基于主从多链的数据分类分级访问控制模型,实现数据的分类分级保障与动态安全访问。首先,构建链上链下混合式可信存储模型,以平衡区块链面临的存储...为解决数据混合存储导致精准查找速度慢、数据未分类分级管理造成安全治理难等问题,构建基于主从多链的数据分类分级访问控制模型,实现数据的分类分级保障与动态安全访问。首先,构建链上链下混合式可信存储模型,以平衡区块链面临的存储瓶颈问题;其次,提出主从多链架构,并设计智能合约,将不同隐私程度的数据自动存储于从链;最后,以基于角色的访问控制为基础,构建基于主从多链与策略分级的访问控制(MCLP-RBAC)机制并给出具体访问控制流程设计。在分级访问控制策略下,所提模型的吞吐量稳定在360 TPS(Transactions Per Second)左右。与BC-BLPM方案相比,发送速率与吞吐量之比达到1∶1,具有一定优越性;与无访问策略相比,内存消耗降低35.29%;与传统单链结构相比,内存消耗平均降低52.03%;与数据全部上链的方案相比,平均存储空间缩小36.32%。实验结果表明,所提模型能有效降低存储负担,实现分级安全访问,具有高扩展性,适用于多分类数据的管理。展开更多
文摘With the development of cloud computing, the mutual understandability among distributed data access control has become an important issue in the security field of cloud computing. To ensure security, confidentiality and fine-grained data access control of Cloud Data Storage (CDS) environment, we proposed Multi-Agent System (MAS) architecture. This architecture consists of two agents: Cloud Service Provider Agent (CSPA) and Cloud Data Confidentiality Agent (CDConA). CSPA provides a graphical interface to the cloud user that facilitates the access to the services offered by the system. CDConA provides each cloud user by definition and enforcement expressive and flexible access structure as a logic formula over cloud data file attributes. This new access control is named as Formula-Based Cloud Data Access Control (FCDAC). Our proposed FCDAC based on MAS architecture consists of four layers: interface layer, existing access control layer, proposed FCDAC layer and CDS layer as well as four types of entities of Cloud Service Provider (CSP), cloud users, knowledge base and confidentiality policy roles. FCDAC, it’s an access policy determined by our MAS architecture, not by the CSPs. A prototype of our proposed FCDAC scheme is implemented using the Java Agent Development Framework Security (JADE-S). Our results in the practical scenario defined formally in this paper, show the Round Trip Time (RTT) for an agent to travel in our system and measured by the times required for an agent to travel around different number of cloud users before and after implementing FCDAC.
文摘Big data has a strong demand for a network infrastructure with the capability to support data sharing and retrieval efficiently. Information-centric networking （ICN） is an emerging approach to satisfy this demand, where big data is cached ubiquitously in the network and retrieved using data names. However, existing authentication and authorization schemes rely mostly on centralized servers to provide certification and mediation services for data retrieval. This causes considerable traffic overhead for the secure distributed sharing of data. To solve this problem, we employ identity-based cryptography （IBC） to propose a Distributed Authentication and Authorization Scheme （DAAS）, where an identity-based signature （IBS） is used to achieve distributed verifications of the identities of publishers and users. Moreover, Ciphertext-Policy Attribnte-based encryption （CP-ABE） is used to enable the distributed and fine-grained authorization. DAAS consists of three phases： initialization, secure data publication, and secure data retrieval, which seamlessly integrate authentication and authorization with the in- terest/data communication paradigm in ICN. In particular, we propose trustworthy registration and Network Operator and Authority Manifest （NOAM） dissemination to provide initial secure registration and enable efficient authentication for global data retrieval. Meanwhile, Attribute Manifest （AM） distribution coupled with automatic attribute update is proposed to reduce the cost of attribute retrieval. We examine the performance of the proposed DAAS, which shows that it can achieve a lower bandwidth cost than existing schemes.
文摘为解决数据混合存储导致精准查找速度慢、数据未分类分级管理造成安全治理难等问题,构建基于主从多链的数据分类分级访问控制模型,实现数据的分类分级保障与动态安全访问。首先,构建链上链下混合式可信存储模型,以平衡区块链面临的存储瓶颈问题;其次,提出主从多链架构,并设计智能合约,将不同隐私程度的数据自动存储于从链;最后,以基于角色的访问控制为基础,构建基于主从多链与策略分级的访问控制(MCLP-RBAC)机制并给出具体访问控制流程设计。在分级访问控制策略下,所提模型的吞吐量稳定在360 TPS(Transactions Per Second)左右。与BC-BLPM方案相比,发送速率与吞吐量之比达到1∶1,具有一定优越性;与无访问策略相比,内存消耗降低35.29%;与传统单链结构相比,内存消耗平均降低52.03%;与数据全部上链的方案相比,平均存储空间缩小36.32%。实验结果表明,所提模型能有效降低存储负担,实现分级安全访问,具有高扩展性,适用于多分类数据的管理。
