A Comprehensive Cloud Security Model with Enhanced Key Management, Access Control and Data Anonymization Features

Abstract

A disgusting problem in public cloud is to securely share data based on fine grained access control policies and unauthorized key management. Existing approaches to encrypt policies and data with different keys based on public key cryptosystem are Attribute Based Encryption and proxy re-encryption. The weakness behind approaches is: It cannot efficiently handle policy changes and also problem in user revocation and attribute identification.  Even though it is so popular, when employed in cloud it generate high computational and storage cost. More importantly, image encryption is some out complex in case of public key cryptosystem. On the publication of sensitive dataset, it does not preserve privacy of an individual. A direct application of a symmetric key cryptosystem, where users are served based on the policies they satisfy and unique keys are generated by Data Owner (DO). Based on this idea, we formalize a new key management scheme, called Symmetric Chaos Based key Management (SCBKM), and then give a secure construction of a SCBKM scheme called AS-Chaos. The idea is to give some secrets to Key Manager (KM) based on the identity attributes they have and later allow them to derive actual symmetric keys based on their secrets. Using our SCBKM construct, we propose an efficient approach for fine-grained encryption-based access control for data stored in untrusted cloud storage.