TL;DR
This paper introduces two new "pairing-free" cryptographic identification schemes that enhance security and efficiency for identity-based systems. These schemes allow for instant revocation of user keys, a crucial feature for access control, by requiring a trusted third party (a security mediator) to participate in the identification process.
What Problem Does This Paper Address?
Traditional public-key cryptography relies on certificates to bind a user's identity to their public key. Managing these certificates becomes complex and costly as the number of users grows. Identity-based cryptography simplifies this by using the user's identity directly as their public key, eliminating the need for certificates. However, a significant challenge in identity-based systems is revoking a user's access quickly and efficiently. Existing methods for key revocation can be slow and resource-intensive. Furthermore, a recent advancement, security-mediated identity-based identification (SM-IBI), introduced a way to achieve instant revocation but relied on "bilinear pairings," which are computationally expensive. This paper addresses the need for efficient, pairing-free SM-IBI schemes that maintain provable security.
Methods
The authors propose two novel SM-IBI schemes:
- GQ-SM-IBI: This scheme is based on the well-established RSA (Rivest–Shamir–Adleman) cryptosystem and the Guillou-Quisquater identification scheme. It leverages the RSA Inversion (RSAI) and One-More RSA Inversion (OMRSAI) assumptions for security proofs.
- BNN-SM-IBI: This scheme is built upon the discrete logarithm (DL) assumption and the BNN identity-based identification scheme. Its security is proven using the Discrete Logarithm (DL) and One-More Discrete Logarithm (OMDL) assumptions.
Both schemes follow the SM-IBI paradigm by splitting a user's secret key into two parts: one held by the user and the other by a security mediator. For any identification process, both parties must cooperate. This cooperation allows the security mediator to maintain a revocation list and deny participation to revoked users, thus enabling instant revocation. The key innovation here is achieving this functionality without using computationally expensive bilinear pairings.
Datasets
Not discussed in the paper.
Key Insights and Ideas
- Pairing-Free SM-IBI is Achievable and Beneficial: The paper demonstrates that the desirable features of security-mediated identity-based identification (instant revocation, identity-based keys) can be achieved without the computational overhead of bilinear pairings, making these schemes more practical.
- Leveraging Existing Cryptographic Primitives: The authors successfully adapt existing, well-understood identification schemes (GQ and BNN) to the security-mediated framework, building on established security assumptions (RSA and DL). This approach lends credibility and robustness to their new schemes.
- The Role of the Security Mediator: The core idea of the security mediator is elegantly integrated. The mediator doesn't just hold a revocation list; their active participation is cryptographically required to form the complete secret key for identification, providing a strong mechanism for control.
- Provable Security: The schemes are rigorously proven secure in the random oracle model against both passive and active/concurrent impersonation attacks, under standard cryptographic assumptions. This is crucial for building trust in cryptographic systems.
Key Results
The paper presents two new SM-IBI schemes that are:
- Pairing-Free: This significantly improves efficiency compared to previous SM-IBI schemes that relied on bilinear pairings.
- Efficient: The paper provides detailed operational and communication cost analyses, showing that their schemes are substantially faster than the existing pairing-based SM-IBI scheme. Table 5, in particular, shows that the GQ-SM-IBI and BNN-SM-IBI schemes have significantly lower identification protocol runtimes, outperforming the pairing-based BLS-SM-IBI by factors of 6-7.
- Provably Secure: The schemes are proven secure against impersonation attacks under well-known cryptographic assumptions (RSA and Discrete Logarithm).
Limitations and Future Work
- Trusted Mediator: A fundamental limitation of security-mediated cryptography, as acknowledged by the authors, is the reliance on a trusted third party (the security mediator). The security of the system hinges on the mediator's integrity and availability.
- Simulation-Based Efficiency: While the paper provides theoretical analysis and simulation results for efficiency, real-world performance can vary based on implementation details and hardware.
- Future Work: The authors suggest that further research could explore other pairing-free cryptographic primitives within the security-mediated framework. Additionally, investigating schemes with even greater efficiency or enhanced security features (e.g., resistance to more sophisticated attacks) would be valuable.
Summary updated