Towards Secure Identity-Based Cryptosystems for Cloud Computing

Hassan Al-Jumyli, Waleed Khalid (2020) Towards Secure Identity-Based Cryptosystems for Cloud Computing. Doctoral thesis, University of Buckingham.

1305979 Towards Secure Identity-Based Crytosystems for Cloud Computing.pdf
Available under License Creative Commons Attribution Non-commercial No Derivatives.

Download (5MB) | Preview


The convenience provided by cloud computing has led to an increasing trend of many business organizations, government agencies and individual customers to migrate their services and data into cloud environments. However, once clients’ data is migrated to the cloud, the overall security control will be immediately shifted from data owners to the hands of service providers. When data owners decide to use the cloud environment, they rely entirely on third parties to make decisions about their data and, therefore, the main challenge is how to guarantee that the data is accessible by data owners and authorized users only. Remote user authentication to cloud services is traditionally achieved using a combination of ID cards and passwords/PINs while public key infrastructure and symmetric key encryptions are still the most common techniques for enforcing data security despite the missing link between the identity of data owners and the cryptographic keys. Furthermore, the key management in terms of the generation, distribution, and storage are still open challenges to traditional public-key systems. Identity-Based Cryptosystems (IBCs) are new generations of public key encryptions that can potentially solve the problems associated with key distribution in public key infrastructure in addition to providing a clear link between encryption keys and the identities of data owners. In IBCs, the need for pre-distributed keys before any encryption/decryption will be illuminated, which gives a great deal of flexibility required in an environment such as the cloud. Fuzzy identity-based cryptosystems are promising extensions of IBCs that rely on biometric modalities in generating the encryption and decryption keys instead of traditional identities such as email addresses. This thesis argues that the adoption of fuzzy identity-based cryptosystems seems an ideal option to secure cloud computing after addressing a number of vulnerabilities related to user verification, key generation, and key validation stages. The thesis is mainly concerned with enhancing the security and the privacy of fuzzy identity-based cryptosystems by proposing a framework with multiple security layers. The main contributions of the thesis can be summarised as follows. 1. Improving user verification based on using a Challenge-Response Multifactor Biometric Authentication (CR-MFBA) in fuzzy identity-based cryptosystems that reduce the impacts of impersonators attacks. 2. Reducing the dominance of the “trusted authority” in traditional fuzzy identity-based cryptosystems by making the process of generating the decryption keys a cooperative process between the trusted authority server and data owners. This leads to shifting control over the stored encrypted data from the trusted authority to the data owners. 3. Proposing a key-validity method that relies on employing the Shamir Secret Sharing, which also contributes to giving data owners more control over their data. 4. Further improving the control of data owners in fuzzy identity-based cryptosystems by linking the decryption keys parameters with their biometric modalities. 5. Proposing a new asymmetric key exchange protocol based on utilizing the scheme of fuzzy identity-based cryptosystems to shared encrypted data stored on cloud computing.

Item Type: Thesis (Doctoral)
Uncontrolled Keywords: Cloud Computing ; Cryptosystems ; User Authentication ; Computer Security
Subjects: Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Q Science > QA Mathematics > QA76 Computer software
Divisions: School of Computing
Depositing User: Nicola Button
Date Deposited: 03 Mar 2022 12:13
Last Modified: 03 Mar 2022 12:13

Actions (login required)

View Item View Item