Research On Key Technologies Of Data Sharing And Repairing In Cloud Computing Environment

Cloud computing, as a promising computing pattern, provides organizations and individuals efficient, inexpensive and professional information services and related infrastructure. Cloud storage is one of the most popular applications supported by cloud computing platform. However, the security issues of the outsourced data in cloud storage figures out to be the most striking problems. In addition, it is increasingly popular for people to access the cloud service with mobile devices with the great development of the intelligence of mobile devices. Unfortunately, the mobile devices are limited with low computation and communication capability as well as battery power compared to traditional PC devices. As a matter of fact, the research on the data sharing system in mobile cloud computing environment is extremely urgent. On the other hand, cooperative work based on cloud is more and more promising with the intensive application of cloud storage. In this case, multiple data owners share their outsourced data to the work group for cooperative purposes. To protect the privacy of the outsoueced data, it is important to study the group-oriented data sharing in cloud storage. Moreover, the integrity and availability of the outsourced data catch more and more attentions since the scale of data stored in the cloud computing platform becomes larger. It is essential for cloud service providers to work out an efficient and secure mechanism to recover the outsourced data in the case of storage corruptions. This dissertation addresses the above challenges and has the following three contributions.1. We put forward a new solution, referred to as flexible access control with outsourceable revocation (FACOR) for mobile clouds. The FACOR applies the attribute-based encryption to enable flexible access control on outsourced data, and allows mobile users to outsource the time-consuming encryption and decryption computations to proxies, with only requiring attributes authorization to be fully trusted. As an advantageous feature, FACOR provides an outsourceable revocation for mobile users to reduce the complicated attribute-based revocation operations. The security analysis shows that our FACOR scheme achieves data security against collusion attacks and unauthorized accesses from revoked users. Both theoretical and experimental results confirm that our proposed scheme greatly reliefs the mobile devices from heavy encryption and decryption computations, as well as the complicated revocation of access rights in mobile clouds.2. We put forward a new paradigm, referred to as secure, scalable and efficient multi-owner (SSEM) data sharing in clouds. The SSEM integrates identity-based encryption and asymmetric group key agreement to enable group-oriented access control for data owners in a many-to-many sharing pattern. Moreover, with SSEM, users can join in or leave from the group conveniently with the privacy of both group data and user data. We proposed the key-ciphertext homomorphism technique to construct an SSEM scheme with short ciphertexts. The security analysis shows that our SSEM scheme achieves data security against unauthorized accesses and collusion attacks. Both theoretical and experimental results confirm that our proposed scheme takes users little costs to share and access outsourced data in a group manner.3. We put forward a new data repairing paradigm, referred to as efficient and secure data repairing (ESDR) paradigm. In ESDR, repairing data is distributed to other storage units after being preprocessed and is collected and reassembled onto the corrupted unit. Following this paradigm, we propose an ESDR scheme by using regenerating code and certificateless signcryption technique. Furthermore, the proposed enc2-mac-signcrypt preprocessing promotes the security and efficiency notably. With the sophisticated design, the confidentiality, integrity and authenticity (CIA) of data can be well maintained to counter threats as wiretapping, tampering and pollution attacks. Both theoretical analysis and experimental evaluation confirm that this scheme is practical and efficient to secure the data repairing in cloud storage.