Sponsored Group Signature and its Application to Privacy-preserving Guest Access in Smart Environments

Technical Summary: Sponsored Group Signature and its Application to Privacy-preserving Guest Access in Smart Environments

Problem Statement
The paper addresses the challenge of providing privacy-preserving, temporary access for guests in smart environments (e.g., smart buildings, universities, corporations). Existing solutions often rely on a centralized Group Manager (GM) that must be online to issue credentials, or they use Ring Signatures which lack accountability. Specifically, standard Group Signatures require a trusted GM for user enrollment, creating a bottleneck for spontaneous guest access. Conversely, solutions that remove the GM (like Ring Signatures) often lack the ability to trace misbehavior or enforce access policies (e.g., limiting the number of tokens per host). The authors identify a need for a mechanism that allows existing members (hosts) to enroll new members (guests) without GM interaction, while maintaining anonymity for the guests and ensuring accountability for the hosts.

Methodology
The authors propose a novel cryptographic primitive called Sponsored Group Signature (SPGS). SPGS is a hierarchical, two-level group signature scheme:

1. Level 1 (Sponsors): These are the original group members (e.g., building residents/hosts). They join the system via interaction with a trusted Group Manager (GM).
2. Level 2 (Sponsored Members): These are new members (e.g., guests) enrolled by Sponsors without interacting with the GM.

Key Properties of SPGS:

• Decentralized Join: Sponsors can "sponsor" new members, removing the need for the GM to be online for every new enrollment.
• Accountability: If a sponsored member misbehaves, their signature can be "opened" by the GM to reveal the identity of the sponsor who enrolled them.
• Linkability: While sponsors enjoy full anonymity (unlinkable signatures), signatures from sponsored members are linkable. This allows the system to track the behavior of a specific guest or count the number of tokens issued by a specific host without revealing their real identities.
• Security: The scheme ensures correctness, non-frameability (users cannot be framed), sponsor anonymity, sponsor traceability, sponsored-member privacy, and sponsored-member linkability.

Construction
The paper provides a generic construction of SPGS in the standard model using three building blocks:

1. A (partially) dynamic Group Signature (GS) scheme.
2. A Commitment Scheme (specifically a homomorphic one).
3. A Knowledge-Sound Non-Interactive Zero-Knowledge (NIZK) proof of knowledge.

In this construction:

• Sponsors interact with the GM to obtain standard group keys.
• Sponsored members generate a random private key and a commitment to it as their public key.
• The sponsor signs this public key using the underlying Group Signature scheme.
• The sponsored member signs messages using a NIZK proof that demonstrates knowledge of the private key corresponding to the committed public key, while treating the message as part of the public parameters.

Application: k-times Anonymous Guest Access Token (k-AGAT)
To demonstrate practical utility, the authors apply SPGS to construct k-AGAT for smart buildings.

• Mechanism: A host acts as a sponsor, and a guest acts as a sponsored member. The host issues a token containing the guest's public key and policy constraints (e.g., expiry time).
• Linkability for Policy Enforcement: The scheme uses pseudonyms to allow the verifier (Access Management Authority) to link multiple tokens to the same host. This enables the system to enforce a "k" limit, ensuring a host cannot issue more than a specific number of guest tokens.
• Traceability: In cases of misbehavior, the Infrastructure Management (IM) can open the token to reveal the hosting entity, ensuring accountability.
• Privacy: The identities of both the host and the guest remain hidden from the verifier during normal operation.

Results and Evaluation
The authors implemented the cryptographic components of the protocol (using BBS group signatures, Pedersen commitments, and Schnorr NIZKs) and compared them against a non-private baseline using standard digital signatures.

• Performance: The implementation shows that while the proposed scheme introduces computational overhead (particularly for the host and IM due to group signature operations and encryption), the costs are in the millisecond range.
• Storage: The storage overhead is manageable for modern smart devices, though the IM requires more storage to maintain member keys and pseudonym lists.
• Feasibility: The results confirm that the privacy-preserving approach is efficient enough for real-world deployment in smart environments, balancing security and performance.

Significance and Claims
The paper claims that SPGS fills a critical gap in privacy-preserving access control by enabling decentralized user join without sacrificing accountability.

• It offers a flexible alternative to traditional Group Signatures where the GM must be online for every join operation.
• It improves upon Ring Signatures and Linkable Spontaneous Anonymous Group (LSAG) signatures by providing a mechanism to trace misbehavior back to a responsible entity (the sponsor) rather than offering unconditional anonymity that prevents accountability.
• The k-AGAT application demonstrates a concrete solution for "zero-touch" guest enrollment in smart buildings, allowing hosts to manage guest access policies (duration, frequency) while preserving the anonymity of both parties from the infrastructure provider.

The authors conclude that their work provides a formal framework and a practical, efficient instantiation for privacy-preserving, accountable guest access in smart environments, with future work suggested for multi-building scenarios and multi-level guest hierarchies.