Blockchain technologies have drawn a lot of attentions, and its immutability is paramount to applications requiring persistent records. However, tremendous real-world incidents have exposed the harm of strict immutability, such as the illicit data stored on Bitcoin and the loss of millions of dollars in vulnerable smart contracts. Moreover, “Right to be Forgotten” has been imposed in new General Data Protection Regulation (GDPR) of European Union, which is incompatible with blockchain's immutability. Therefore, it is imperative to design efficient redactable blockchain in a controlled way. In this paper, we present a generic design of redactable blockchain protocols in the permissionless setting, applied to both proof-of-stake and proof-of-work blockchains. Our protocol can (1) maintain the same adversary bound requirement as the underlying blockchain, (2) support various network environments, (3) offer public verifiability for any redaction, and (4) achieve instant redaction, even only within one slot in the best case, which is desirable for redacting harmful data. Furthermore, we define the first ideal protocol of redactable blockchain and conduct security analysis following the language of universal composition. Finally, we develop a proof-of-concept implementation showing that the overhead remains minimal for both online and re-spawning nodes, which demonstrates the high efficiency of our design.

Data availability is the major concern of outsourced data storage services. While reliability is promoted by many storage providers, it is infeasible for users to verify these claims. Proofs of Storage-Time are proposed to address this issue: they allow a prover to convince a verifier that the prover indeed stores the outsourced data continuously during the whole storage period. These protocols, however, either fail to guarantee the actual duration of data possession, or require the client to perform a computationally expensive storing process for each file, marking them far from being practical. We present$\mathsf {ePoSt}$, the first secure, stateless and efficient Proof of Storage-Time protocol with public verifiability. Not only does it ensure continuous data availability, but it also minimizes the cost of the client, which is crucial for real-world deployment. Specifically, processing a 1GB file for outsourcing, in anticipation for 3 proof/verify interactions, each attesting a period of 1 year at 25 minutes intervals requires only 66.34 minutes. In contrast, state-of-the-art solution requires 51 hours of processing to prepare a file for outsourcing. Proof size and verification remains at a reasonable cost: a proof is of size 48.82KB, and can be verified in 36.05 ms. Furthermore, our solution enjoys public verifiability and remains stateless. The former allows the data owner to outsource the verification process, and the latter allows unlimited number of proofs and verification to be conducted after the file has been stored. These properties make$\mathsf {ePoSt}$particularly suitable for new business models such as decentralised storage networks (e.g., Filecoin).

OAuth 2.0 is one of the most widely used Internet protocols for authorization/single sign-on (SSO) and is also the foundation of the new SSO protocol OpenID Connect. Due to its complexity and its flexibility, it is difficult to comprehensively analyze the security of the OAuth 2.0 standard, yet it is critical to obtain practical security guarantees for OAuth 2.0. In this paper, we present the first computationally sound security analysis of OAuth 2.0. First, we introduce a new primitive, the three-party authenticated secret distribution (3P-ASD for short) protocol, which plays the role of issuing the secret and captures the token issue process of OAuth 2.0. As far as we know, this is the first attempt to formally abstract the authorization technology into a general primitive and then define its security. Then, we present a sufficiently rich three-party security model for OAuth protocols, covering all kinds of authorization flows, providing reasonably strong security guarantees and moreover capturing various web features. To confirm the soundness of our model, we also identify the known attacks against OAuth 2.0 in the model. Furthermore, we prove that two main modes of OAuth 2.0 can achieve our desired security by abstracting the token issue process into a 3P-ASD protocol. Our analysis is not only modular which can reflect the compositional nature of OAuth 2.0, but also fine-grained which can evaluate how the intermediate parameters affect the final security of OAuth 2.0.

Proof-of-stake blockchain protocols are becoming one of the most promising alternatives to the energy-consuming proof-of-work protocols. However, one particularly critical threat in the PoS setting is the well-known long-range attacks caused by secret key leakage (LRSL attack). Specifically, an adversary can attempt to control/compromise accounts possessing substantial stake at some past moment such that double-spend or erase past transactions, violating the fundamental persistence property of blockchain. Puncturable signatures provide a satisfying solution to construct practical proof-of-stake blockchain resilient to LRSL attack, despite of the fact that existent constructions are not efficient enough for practical deployments. In this paper, we provide an in-depth study of puncturable signatures and explore its applications in the proof-of-stake blockchain. We formalize a security model that allows the adversary for adaptive signing and puncturing queries, and show a construction with efficient puncturing operations based on the Bloom filter data structure and strong Diffie-Hellman assumption. The puncturing functionality we desire is for a particular part of message, like prefix, instead of the whole message. Furthermore, we use puncturable signatures to construct practical proof-of-stake blockchain protocols that are resilient to LRSL attack, while previously the forward-secure signature is used to immunize this attack. We implement our scheme and provide experimental results showing that in comparison with the forward-secure signature, our construction performs substantially better on signature size, signing and verification efficiency, significantly on key update efficiency.

The Transport Layer Security (TLS) protocol is by far the most widely deployed protocol for securing communications and the Internet Engineering Task Force (IETF) is currently developing TLS 1.3 as the next-generation TLS protocol. The TLS standard features multiple modes of handshake protocols and supports many combinational running of successive TLS handshakes over multiple connections. Although each handshake mode is now well-understood in isolation, their composition in TLS 1.2 remains problematic, and yet it is critical to obtain practical security guarantees for TLS. In this paper, we present the first formal treatment of multiple handshakes protocols of TLS 1.3 candidates. First, we introduce a multi-level&stage security model, an adaptation of the BellareRogaway authenticated key exchange model, covering all kinds of compositional interactions between different TLS handshake modes and providing reasonably strong security guarantees. Next, we prove that candidate handshakes of TLS 1.3 draft meet our strong notion of multiple handshakes security. Our results confirm the soundness of TLS 1.3 security protection design. Such a multi-level&stage approach is convenient for analyzing the compositional design of the candidates with different session modes, as they establish dependencies of multiple sessions. We also identify the triple handshake attack of Bhargavan et al. on TLS 1.2 within our multiple handshakes security model. We show generically that the proposed fixes (RFC 7627) for TLS 1.2 offer good protection against multiple handshakes attacks.