Biblio
Outsourcing data storage and IT workloads to a third-party cloud provider introduces some security risks and time performance degradation. Moreover, controlling access to this data becomes very difficult when the volume of the data and number of users is very high. Various access control techniques have been proposed to address this issue. However, those techniques have complex schemes which are costly to be applied in real scenarios and they have limited flexibility and scalability to large volumes of data and users. In this paper we propose ESSAC which is an enhanced version of the SSAC scheme. ESSAC introduces a fine-grained access control scheme based on a classified Attribute Based Encryption, Role Based Encryption and Single Key Encryption methodology which achieves highest security without degrading the performance. We validate our scheme using a simulation on top of Amazon S3 and compare it to current schemes.
Delay-Tolerant Networks exhibit highly asynchronous connections often routed over many mobile hops before reaching its intended destination. The Bundle Security Protocol has been standardized providing properties such as authenticity, integrity, and confidentiality of bundles using traditional Public-Key Cryptography. Other protocols based on Identity-Based Cryptography have been proposed to reduce the key distribution overhead. However, in both schemes, secret keys are usually valid for several months. Thus, a secret key extracted from a compromised node allows for decryption of past communications since its creation. We solve this problem and propose the first forward secure protocol for Delay-Tolerant Networking. For this, we apply the Puncturable Encryption construction designed by Green and Miers, integrate it into the Bundle Security Protocol and adapt its parameters for different highly asynchronous scenarios. Finally, we provide performance measurements and discuss their impact.