The last decade has witnessed a sharp spike in the popularity of group applications, such as multi-media conferencing, collaborative workspaces, as well as dynamic coalitions common in law enforcement and disaster rescue scenarios. Standard security services required in such group settings (such as confidentiality, integrity and authentication) can be efficiently achieved if all group members share a secret key. A group key agreement protocol (GKA) provide a technique for any set of participating players to collaboratively establish a shared key.
In this thesis, we tackle two challenging issues in GKA: robustness and privacy-preserving authentication. Robustness improves performance by guaranteeing completion of GKA protocols despite participating player or communication failures. Affiliation-hiding authenticated GKA prevents from man-in-the-middle-attacks by authenticating players in a privacy-preserving manner.
We propose a two-round group key agreement protocol, which offers a natural trade-off between message size and the desired level of fault-tolerance. We demonstrate that in a realistic setting of random player faults, the new protocol has logarithmic-sized messages and expected round complexity close to two. We generalize a robust group key agreement protocol to support heterogeneous players for a general environment. We prove the security of the proposed protocols under the standard cryptographic assumptions.
We present the formal definition of the affiliation-hiding GKA in a standard PKI model with re-usable certificates. The model includes strong security properties, called perfect forward secrecy (PFS) and linkable affiliation-hiding (LAH)—that an key agreement protocol might provide in the face of player corruptions and session compromises. We construct two variants of optimal-cost AH-AGKA protocols, which satisfy PFS and LAH notions under the standard cryptographic assumptions in the random oracle model. The protocol yields minimal costs: three rounds of interaction matches the round complexity of non-private PFS authenticated GKA and its computation costs are comparable to the cost of unauthenticated GKA.
The last decade has witnessed a sharp spike in the popularity of group applications, such as multi-media conferencing, collaborative workspaces, as well as dynamic coalitions common in law enforcement and disaster rescue scenarios. Standard security services required in such group settings (such as confidentiality, integrity and authentication) can be efficiently achieved if all group members share a secret key. A group key agreement protocol (GKA) provide a technique for any set of participating players to collaboratively establish a shared key.
In this thesis, we tackle two challenging issues in GKA: robustness and privacy-preserving authentication. Robustness improves performance by guaranteeing completion of GKA protocols despite participating player or communication failures. Affiliation-hiding authenticated GKA prevents from man-in-the-middle-attacks by authenticating players in a privacy-preserving manner.
We propose a two-round group key agreement protocol, which offers a natural trade-off between message size and the desired level of fault-tolerance. We demonstrate that in a realistic setting of random player faults, the new protocol has logarithmic-sized messages and expected round complexity close to two. We generalize a robust group key agreement protocol to support heterogeneous players for a general environment. We prove the security of the proposed protocols under the standard cryptographic assumptions.
We present the formal definition of the affiliation-hiding GKA in a standard PKI model with re-usable certificates. The model includes strong security properties, called perfect forward secrecy (PFS) and linkable affiliation-hiding (LAH)—that an key agreement protocol might provide in the face of player corruptions and session compromises. We construct two variants of optimal-cost AH-AGKA protocols, which satisfy PFS and LAH notions under the standard cryptographic assumptions in the random oracle model. The protocol yields minimal costs: three rounds of interaction matches the round complexity of non-private PFS authenticated GKA and its computation costs are comparable to the cost of unauthenticated GKA.
A robust group key agreement protocol (GKA) allows a set of players to establish a shared secret key, regardless of network/node failures. Current constant-round GKA protocols are either efficient and nonrobust or robust but not efficient; assuming a reliable broadcast communication medium, the standard encryption-based group key agreement protocol can be robust against arbitrary number of node faults, but the size of the messages broadcast by every player is proportional to the number of players. In contrast, nonrobust group key agreement can be achieved with each player broadcasting just constant-sized messages. We propose a novel 2-round group key agreement protocol, which tolerates up to T node failures, using O(T)-sized messages for any T. We show that the new protocol implies a fully-robust group key agreement with logarithmic-sized messages and expected round complexity close to 2, assuming random node faults. The protocol can be extended to withstand malicious insiders at small constant factor increases in bandwidth and computation. The proposed protocol is secure under the (standard) Decisional Square Diffie-Hellman assumption.
Abstract. Privacy concerns in many aspects of electronic communication trigger the need to re-examine – with privacy in mind – familiar security services, such as authentication and key agreement. An Affiliation-Hiding Group Key Agreement (AH-AGKA) protocol (also known as Group Secret Handshake) allows a set of participants, each with a certificate issued by the same authority, to establish a common authenticated secret key. In contrast to standard AGKA protocols, an AH-AGKA protocol has the following privacy feature: If Alice, who is a member of a group G, participates in an AH-AGKA protocol, none of the other protocol participants learn whether Alice is a member of G, unless these participants are themselves members of group G. Such protocols are useful in suspicious settings where a set of members of a (perhaps secret) group need to authenticate each other and agree on a common secret key, without revealing their affiliations to outsiders. In this paper we strengthen the prior definition of AH-AGKA so that the security and privacy properties are maintained under any composition of protocol instances. We also construct two novel AH-AGKA protocols secure in this new and stronger model under the RSA and Gap Diffie-Hellman assumptions, respectively. Each protocol involves only two communication rounds and few exponentiations per player (e.g., no bilinear map operations). Interestingly, these costs are essentially the same as those of the underlying (unauthenticated) group key agreement protocol. Finally, our protocols, unlike prior results, retain their security and privacy properties without the use of one-time certificates.
We introduce the concept of Affiliation-Hiding Authenticated Asymmetric Group Key Agreement (AH-AAGKA) and construct a concrete one-round AH-AAGKA protocol. An AH-AAGKA protocol allows the participants of a group to establish a common encryption key associated with several decryption keys; each of which can only be computed by the corresponding legitimate group member. An AH-AAGKA protocol has the following privacy feature. For a member 𝒰i of a group G, if 𝒰i participates in an AH-AAGKA protocol, any protocol participant 𝒰j cannot learn whether 𝒰i is a member of G, unless 𝒰j himself is a member of group G. Our scheme demonstrates new features in comparison with other existing AH-AGKA protocols. If non-group members participate in our protocol, honest parties can identify these non-group members. Our scheme also captures Unlinkability and Perfect Forward Secrecy (PFS), which are missing in other existing schemes. We propose a novel security model to prove that our protocol holds PFS and present a new privacy model to prove that our scheme meets Affiliation-Hiding property.
The use of bilinear pairings as a building block for cryptographic protocols, most notably in the construction of identity-based cryptosystems, is a very popular area of cryptographic research. In this thesis, we provide a novel classification of pairing-based group key agreement (GKA) from current literature. We propose a new framework for constructing secure and efficient computationally asymmetric authenticated GKA protocols from identity-based signcryption schemes and adapt this framework to construct a novel identity-based authenticated GKA protocol with perfect forward secrecy. To the best of our knowledge, our protocol is the first that maintains perfect forward secrecy in the presence of auxiliary key agreement protocols. We formally prove the security of our protocols in the random oracle model and show that they are communication and computationally efficient in comparison to the pairing-based protocols from the literature.
In modern collaborative and distributed applications, authenticated group key agreement (AGKA) is one of the important issues. In this paper, we present that one existing certificateless AGKA protocol has some security drawbacks and does not satisfy some security attributes required for AGKA protocols. To overcome its drawbacks, we propose a secure and efficient protocol upon the initial one and a signature scheme proved secure in the random oracle model. The security analysis shows that our improved protocol can achieve Explicit key authentication, Known session key security, Known ephemeral secrets security, Perfect forward security, No key compromise impersonation, No key control and No unknown key share. With respect to efficiency, our protocol only needs three communication rounds. In each protocol run, only three pairing computations are involved for each participant.