Secure Data Sharing and Collaboration in the Cloud

Cloud technology can be leveraged to enable data-sharing capabilities, which can benefit the user through greater productivity and efficiency. However, the Cloud is susceptible to many privacy and security vulnerabilities, which hinders the progress and widescale adoption of data sharing for the purposes of collaboration. Thus, there is a strong demand for data owners to not only ensure that their data is kept private and secure in the Cloud, but to also have a degree of control over their own data contents once they are shared with data consumers. Specifically, the main issues for data sharing in the Cloud include key management, security attacks, and data-owner access control. In terms of key management, it is vital that data must first be encrypted before storage in the Cloud, to prevent privacy and security breaches. However, the management of encryption keys is a great challenge. The sharing of keys with data consumers has proven to be ineffective, especially when considering data-consumer revocation. Security attacks may also prevent the widescale usage of the Cloud for data-sharing purposes. Common security attacks include insider attacks, collusion attacks, and man-in-the-middle attacks. In terms of access control, authorised data consumers could do anything they wish with an owner's data, including sending it to their peers and colleagues without the data owner's knowledge. Throughout this thesis, we investigate ways in which to address these issues. We first propose a key partitioning technique that aims to address the key management problem. We deploy this technique in a number of scenarios, such as remote healthcare management. We also develop secure data-sharing protocols that aim to mitigate and prevent security attacks on the Cloud. Finally, we focus on giving the data owner greater control, by developing a self-controlled software object called SafeProtect