Abstract
Each cloud service has numerous owners and tenants, so it is necessary to construct a privacy preserving identity management and access control mechanism for cloud computing. On one hand, cloud service providers (CSP) depend on tenant's identity information to enforce appropriate access control so that cloud resources are only accessed by the authorized tenants who are willing to pay. On the other hand, tenants wish to protect their personalized service access patterns, identity privacy information and accessing newfangled cloud services by on-demand ways within the scope of their permissions. There are many identity authentication and access control schemes to address these challenges to some degree, however, there are still some limitations. In this paper, we propose a new comprehensive approach, called Privacy pReserving Identity and Access Management scheme, referred to as PRIAM, which is able to satisfy all the desirable security requirements in cloud computing. The main contributions of the proposed PRIAM scheme are threefold. First, it leverages blind signature and hash chain to protect tenant's identity privacy and implement secure mutual authentication. Second, it employs the service-level agreements to provide flexible and on-demand access control for both tenants and cloud services. Third, it makes use of the BAN logic to formally verify the correctness of the proposed protocols. As a result, our proposed PRIAM scheme is suitable to cloud computing thanks to its simplicity, correctness, low overhead, and efficiency.
Introduction
Cloud computing has revolutionized the way services are delivered, offering scalability and cost-efficiency. However, this paradigm shift introduces significant privacy concerns. Cloud Service Providers (CSPs) require identity verification to manage access and billing, while tenants demand anonymity to protect their personal data and usage patterns. Existing Identity and Access Management (IAM) solutions often struggle to balance these conflicting needs, frequently failing to provide true "unlinkability" or suffering from high computational overhead.
The PRIAM (Privacy pReserving Identity and Access Management) scheme addresses these gaps by providing a framework that ensures privacy, mutual authentication, and efficient access control.
The PRIAM System Model
The architecture of PRIAM is designed to support a multi-tenant cloud environment where users access services through various network configurations, including local area networks (LANs) and mobile connections.
As shown in the system model, tenants interact with cloud services through an IAM layer. This layer is responsible for mediating access while preserving the tenant's privacy. The challenge lies in ensuring that the IAM can verify the tenant's authorization without necessarily knowing their specific identity or being able to link multiple sessions to the same individual.
The PRIAM Framework and Protocol Flow
The PRIAM framework decomposes the IAM process into several specialized components to handle registration, auditing, and policy enforcement.
The framework consists of the following key entities:
- Registration Server (RS): Handles initial user registration using blind signatures to ensure the server never sees the user's actual identity.
- Audit Service (AS): Monitors transactions for security and compliance.
- Policy Enforcement Point (PEP) and Policy Decision Point (PDP): Manage access requests based on XACML policies and Service-Level Agreements (SLAs).
- Cloud Service Provider (CSP): Hosts the actual resources and interacts with the PDP to grant or deny access.
Core Cryptographic Mechanisms
PRIAM utilizes two primary cryptographic primitives to achieve its goals:
- Blind Signatures: These allow the RS to sign a user's credential "blindly." Much like a notary signing a sealed envelope, the RS verifies the user's right to a credential without learning the credential's content or the user's identity.
- Hash Chains: To avoid the inefficiency of generating new credentials for every login, PRIAM employs hash chains. A user registers once and receives a chain of tokens. Each subsequent login uses the next link in the chain, which is computationally inexpensive to verify.
Security and Efficiency
A critical aspect of PRIAM is its formal verification using BAN (Burrows-Abadi-Needham) logic. This mathematical approach proves that the mutual authentication and key exchange protocols are logically sound and resistant to common attacks like man-in-the-middle or replay attacks.
In terms of performance, PRIAM is highly efficient. During the most frequent phase—pre-authorization—a tenant only needs to perform two public key encryptions and two hash operations. By eliminating heavy exponential calculations during routine access, PRIAM remains suitable for resource-constrained mobile devices.
Conclusion
The PRIAM scheme offers a comprehensive solution for privacy-preserving IAM in the cloud. By integrating blind signatures, hash chains, and SLA-based access control, it achieves a high level of security coverage—satisfying ten distinct security properties—while maintaining low computational overhead. This makes it a robust candidate for modern, privacy-conscious cloud environments.
Summary updated