A multi-party private join (MPPJ) protocol enables multiple source parties to provide a receiver party with the inner joins over their respective datasets, while revealing as little information as possible. There is currently no protocol that directly and efficiently enables such a MPPJ beyond the two- or three-party setting. The presently known protocols either achieve weaker functionality (e.g., multi- party private set intersection protocols) or more general ones (e.g., private-join-compute and generic secure multi-party computation protocols) and are therefore more costly to run for the sources. This work formally introduces MPPJ as an explicit goal, and proposes an efficient, helper-assisted protocol that achieves 𝑛-party inner joins with small leakage and close-to-optimal overhead for the sources. Specifically, for 𝑛 databases with 𝑚 rows, it requires only a single 𝑂 (𝑚) upload from the sources to the helper, and a single 𝑂 (𝑛 · 𝑚) download from the helper to the receiver. Moreover, the helper is entirely oblivious: it enables the efficiency and simplicity goals we are striving for, but it does not learn anything about the computation it facilitates. We formally model and prove the security of our protocol from standard assumptions, in the passive-adversary model. Then, we provide an open-source implementation and an extensive performance evaluation. According to our experiments, our protocol requires 1.02 to 20 times less communication than a current private-join-compute protocol (with no computation over the join) for 2 to 6 parties and input database sizes from 1.5K to 250K records. Finally, we demonstrate the versatility of our approach by extending our protocol to threshold-joins.

Asymmetric Password-Authenticated Key Exchange (aPAKE) enables secure key establishment between a client and a server using a pre-shared password, while providing security against offline attacks. However, aPAKE does not guarantee any precomputation resistance, and considers passwords to become immediately available upon server compromise. A recent work by Dayanikli and Lehmann (EuroS&P'24) observed that many existing aPAKE protocols provide stronger precomputation attack resistance than what is guaranteed through the aPAKE model: they often rely on salted password hashes, where a unique salt makes precomputation attacks more difficult. While these salts are sent in clear to the client during authentication, and thus trivial to obtain for an attacker, this makes a difference in multi-user settings with millions of user accounts per server. In order to run bulk precomputation attacks on all users' passwords, the attacker needs to start an authentication session on behalf of every user to obtain their salts. However, this protection is still limited as salts are static, and the attacker can gradually extract all salt values for precomputation attacks.

Privacy Pass is an anonymous authentication protocol which was initially designed by Davidson et al. (PETS'18) to reduce the number of CAPTCHAs that TOR users must solve. It issues single-use authentication tokens with anonymous and unlinkable redemption guarantees. The issuer and verifier of the protocol share a symmetric key, and tokens are privately verifiable. The protocol has sparked interest from both academia and industry, which led to an Internet Engineering Task Force (IETF) standard. While Davidson et al. formally analyzed the original protocol, the IETF standard introduces several changes to their protocol. Thus, the standardized version's formal security remains unexamined. We fill this gap by analyzing the IETF standard's privately verifiable Privacy Pass protocol.

Single Sign-On (SSO) allows users to conveniently authenticate to many Relying Parties (RPs) through a central Identity Provider (IdP). SSO supports unlinkable authentication towards the RPs via pairwise pseudonyms, where the IdP assigns the user an RP-specific pseudonym. This feature has been rolled out prominently within Apple's SSO service. While establishing unlinkable identities provides privacy towards RPs, it actually emphasizes the main privacy problem of SSO: with every authentication request, the IdP learns the RP that the user wants to access. Solutions to overcome this limitation exist, but either assume users to behave honestly or require them to manage long-term cryptographic keys. In this work, we propose the first SSO system that can provide such pseudonymous authentication in an unobservable yet strongly secure and convenient manner. That is, the IdP blindly derives the user's pairwise pseudonym for the targeted RP without learning the RP's identity and without requiring key material handled by the user. We formally define the desired security and privacy properties for such unlinkable, unobservable, and strongly secure SSO. In particular, our model includes the often neglected RP authentication: the IdP typically wants to limit its services to registered RPs only and thus must be able to (blindly) verify that it issues the token and pseudonym to such a registered RP. We propose a simple construction that combines signatures with efficient proofs-of-knowledge with a blind, yet verifiable, evaluation of the Hashed-Diffie-Hellman PRF. We prove the security of our construction and demonstrate its efficiency through a prototypical implementation, which requires a running time of 2-12ms per involved party.

Password-protected key retrieval (PPKR) enables users to store and retrieve high-entropy keys from a server securely. The process is bootstrapped from a human-memorizable password only, addressing the challenge of how end-users can manage cryptographic key material. The core security requirement is protection against a corrupt server, which should not be able to learn the key or offline- attack it through the password protection. PPKR is deployed at a large scale with the WhatsApp Backup Protocol (WBP), allowing users to access their encrypted messaging history when switching to a new device. Davies et al. (Crypto'23) formally analyzed the WBP, proving that it satisfies most of the desired security. The WBP uses the OPAQUE protocol for password-based key exchange as a building block and relies on the server using a hardware security module (HSM) for most of its protection. In fact, the security analysis assumes that the HSM is incorruptible - rendering most of the heavy cryptography in the WBP obsolete.

This paper analyses the Secure Remote Password Protocol (SRP) in the context of provable security. SRP is an asymmetric Password-Authenticated Key Exchange (aPAKE) protocol introduced in 1998. It allows a client to establish a shared cryptographic key with a server based on a password of potentially low entropy. Although the protocol was part of several standardization efforts, and is deployed in numerous commercial applications such as Apple Homekit, 1Password or Telegram, it still lacks a formal proof of security. This is mainly due to some of the protocol's design choices which were implemented to circumvent patent issues. Our paper gives the first security analysis of SRP in the universal composability (UC) framework. We show that SRP is UC-secure against passive eavesdropping attacks under the standard CDH assumption in the random oracle model. We then highlight a major protocol change designed to thwart active attacks and propose a new assumption – the additive Simultaneous Diffie Hellman (aSDH) assumption – under which we can guarantee security in the presence of an active attacker. Using this new assumption as well as the Gap CDH assumption, we prove security of the SRP protocol against active attacks. Our proof is in the “Angel-based UC framework”, a relaxation of the UC framework which gives all parties access to an oracle with super-polynomial power. In our proof, we assume that all parties have access to a DDH oracle (limited to finite fields). We further discuss the plausibility of this assumption and which level of security can be shown without it.

Light Fidelity (LiFi) networks transmit information via light waves and are an interesting alternative to Radio Frequency networks: as light can be confined easily, LiFi provides better performance and makes eavesdropping attacks much more difficult. A core application of LiFi networks is self-contained and local networks among a group of autonomous devices, e.g., in industrial or medical environments. Cryptographic protocols are used to secure these networks, however the key exchange sometimes relies solely on the confineability of light signals and sends key material in plain over the network. This is clearly not desirable from a security perspective and newer standards recommend key exchange protocols to establish shared keys. A crucial part in any authenticated key exchange protocol is how to bootstrap trust, e.g., by assuming a PKI, pre-installed keys or an out-of-band-channel. Well established solutions exist, but they are not ideal for the type of self-contained networks targeted by LiFi communication. In this work we investigate how the physical properties of a LiFi channel can be used to replace these mechanisms, resulting in a more convenient and also more efficient solution for key exchange. To this end we propose a new type of secret-less key exchange (SEKA) that does not rely on any pre-shared secrets, and instead runs in two phases: a short bootstrap phase where we make stronger assumptions on the physical security, ruling out active attacks. This can be realized by putting all devices in a closed room, taking advantage of the light's confineability feature. The bootstrap phase is followed by a more classical key-exchange phase, where the actual key material gets exchanged in the presence of active attacks - relying on the shared states from the bootstrap phase. We formally define this new type of key-exchange protocol which offers authenticated key exchange with post-compromise security without relying on pre-shared secrets. We then show that a simpler and more efficient version of the signed Diffie-Hellmann protocol, now relying on MACs instead of signatures for the mutual authentication, can be proven secure in our model. Finally, a proof-of-concept implementation of the SEKA protocol is evaluated in a testbed demonstrating the efficiency gains of our approach.

Asymmetric Password-Authenticated Key Exchange (aPAKE) protocols, particularly Strong aPAKE (saPAKE) have enjoyed significant attention, both from academia and industry, with the well-known OPAQUE protocol currently undergoing standardization. In (s)aPAKE, a client and a server collaboratively establish a high-entropy key, relying on a previously exchanged password for authentication. A main feature is its resilience against offline and precomputation (for saPAKE) attacks. OPAQUE, as well as most other aPAKE protocols, have been designed and analyzed in a single-user setting, i.e., modelling that only a single user interacts with the server. By the composition framework of UC, security for the actual multi-user setting is then conjectured. As any real-world (s)aPAKE instantiation will need to cater multiple users, this introduces a dangerous gap in which developers are tasked to extend the single-user protocol securely and in a UC-compliant manner. In this work, we extend the (s)aPAKE definition to directly model the multi-user setting, and explicitly capture the impact that a server compromise has across user accounts. We show that the currently standardized multi-user version of OPAQUE might not provide the expected security, as it is insecure against offline attacks as soon as the file for one user in the system is compromised. This is due to using shared state among different users, which violates the UC composition framework. However, we show that another change introduced in the standardization draft which also involves a shared state does not compromise security. When extending the aPAKE security in the multi-client setting, we notice that the widely used security definition captures significantly weaker security guarantees than what is offered by many protocols. Essentially, the aPAKE definition assumes that the server stores unsalted password-hashes, whereas several protocols explicitly use a salt to protect against precomputation attacks. We therefore propose a definitional framework that captures different salting approaches - thus showing that the security gap between aPAKE and saPAKE can be smaller than expected.

Digital Covid certificates are the first widely deployed end-user cryptographic certificates. For service providers, such as airlines or event ticket vendors, that needed to check that their (global) customers satisfy certain health policies, the verification of such Covid certificates was challenging though - not because of the cryptography involved, but due to the multitude of issuers, different certificate types and the evolving nature of country-specific policies that had to be supported. As Covid certificates contain sensitive health information, their (online) presentation to non-health related entities also poses clear privacy risk. To address both challenges, the EU proposed a specification for outsourcing the verification process to a validator service, that executes the process and informs service providers of the result. The WHO announced to adapt this approach for general vaccination credentials beyond Covid-19. While being beneficial to improve security and privacy for service providers, their solution requires strong trust assumption for the (central) validation service that learns all health-related details of the users. In our work, we propose and formally model a privacy-preserving variant of such an outsourced validation service. Therein the validator learns the attributes it is supposed to verify, but not the users identity. Still, the validator’s assertion is blindly bound to the user’s identity to ensure the desired user-binding. We analyze the EU specification in our model and show that it only meets a subset of those goals. Our analysis further shows that the EU protocol is unnecessarily complex and can be significantly simplified while maintaining the same (weak) level of security. Finally, we propose a new construction for privacy-preserving certificate validation that provably satisfies all desired goals.

OpenID Connect (OIDC) is a Single Sign-On (SSO) protocol that allows users to authenticate to various Relying Parties (RPs) via an Identity Provider (IdP). The main drawback of SSO is its lack of privacy, as the IdP learns the RP’s identity at each user’s login. OIDC supports several protocol flows, of which only one, the Implicit Flow, gives hope for any privacy, as it does not require direct communication between the IdP and RP. This design was initially intended for RPs with technical limitations that prevent them from storing credentials and thus authenticating to the IdP. However, RP authentication is crucial to ensure that users only access properly registered RPs. As a result, the Implicit Flow is being discussed to be excluded from the OAuth specification on which OIDC is based. This paper demonstrates a privacy-preserving approach incorporating RP authentication into the Implicit Flow. The IdP can restrict its service to authenticated RPs and tie each authentication token to a specific user and RP without acquiring knowledge of which user is accessing which RP. We formally define the desired security and privacy properties of such an authenticated Implicit Flow, propose a provably secure construction from generic building blocks, and report on an implementation of our scheme

Cloud storage is becoming increasingly popular among end users that outsource their personal data to services such as Dropbox or Google Drive. For security, uploaded data should ideally be encrypted under a key that is controlled and only known by the user. Current solutions that support user-centric encryption either require the user to manage strong cryptographic keys, or derive keys from weak passwords. While the former has massive usability issues and requires secure storage by the user, the latter approach is more convenient but offers only little security since encrypted data is susceptible to offline attacks. The recent concept of password-authenticated secret-sharing (PASS) enables users to securely derive strong keys from weak passwords by leveraging a distributed server setup, and has been considered a promising step towards secure and usable encryption. However, using PASS for encryption is not as suitable as originally thought: it only considers the (re)construction of a single, static key -- whereas practical encryption will require the management of many, object-specific keys. Using a dedicated PASS instance for every key makes the solution vulnerable against online attacks, inherently leaks access patterns to the servers and poses the risk of permanent data loss when an incorrect password is used at encryption. We therefore propose a new protocol that directly targets the problem of boostrapping encryption from a single password: distributed password-authenticated symmetric encryption DPaSE. DPaSE offers strong security and usability, such as protecting the user's password against online and offline attacks, and ensuring message privacy and ciphertext integrity as long as at least one server is honest. We formally define the desired security properties in the UC framework and propose a provably secure instantiation. The core of our protocol is a new type of Oblivious Pseudorandom Function (OPRF) that allows to extend a previous partially-blind query with a follow-up request and will be used to blindly carry over passwords across evaluations and avoid online attacks. Our (proof-of-concept) implementation of DPaSE uses 10 exponentiations at the user, 4 exponentiations and 2 pairings at each server, and has a server throughput of 76 account creations and 37 (user authentication followed by) encryptions per second, when run between a user and 2-10 servers.

In recent years, oblivious pseudorandom functions (OPRFs) have become a ubiquitous primitive used in cryptographic protocols and privacy-preserving technologies. The growing interest in OPRFs, both theoretical and applied, has produced a vast number of different constructions and functionality variations. In this paper, we provide a systematic overview of how to build and use OPRFs. We first categorize existing OPRFs into essentially four families based on their underlying PRF (Naor-Reingold, Dodis-Yampolskiy, Hashed Diffie-Hellman, and generic constructions). This categorization allows us to give a unified presentation of all oblivious evaluation methods in the literature, and to understand which properties OPRFs can (or cannot) have. We further demonstrate the theoretical and practical power of OPRFs by visualizing them in the landscape of cryptographic primitives, and by providing a comprehensive overview of how OPRFs are leveraged for improving the privacy of internet users. Our work systematizes 15 years of research on OPRFs and provides inspiration for new OPRF constructions and applications thereof.

Single Sign-On (SSO) is becoming an increasingly popular authentication method for users that leverages a trusted Identity Provider (IdP) to bootstrap secure authentication tokens from a single user password. It alleviates some of the worst security issues of passwords, as users no longer need to memorize individual passwords for all service providers, and it removes the burden of these service to properly protect huge password databases. However, SSO also introduces a single point of failure. If compromised, the IdP can impersonate all users and learn their master passwords. To remedy this risk while preserving the advantages of SSO, Agrawal et al. (CCS'18) recently proposed a distributed realization termed PASTA (password-authenticated threshold authentication) which splits the role of the IdP across n servers. While PASTA is a great step forward and guarantees security as long as not all servers are corrupted, it uses a rather inflexible corruption model: servers cannot be corrupted adaptively and - even worse - cannot recover from corruption. The latter is known as proactive security and allows servers to re-share their keys, thereby rendering all previously compromised information useless. In this work, we improve upon the work of PASTA and propose a distributed SSO protocol with proactive and adaptive security (PESTO), guaranteeing security as long as not all servers are compromised at the same time. We prove our scheme secure in the UC framework which is known to provide the best security guarantees for password-based primitives. The core of our protocol are two new primitives we introduce: partially-oblivious distributed PRFs and a class of distributed signature schemes. Both allow for non-interactive refreshing of the secret key material and tolerate adaptive corruptions. We give secure instantiations based on the gap one-more BDH and RSA assumption respectively, leading to a highly efficient 2-round PESTO protocol. We also present an implementation and benchmark of our scheme in Java, realizing OAuth-compatible bearer tokens for SSO, demonstrating the viability of our approach.

Vehicle-to-vehicle (V2V) communication systems are currently being prepared for real-world deployment, but they face strong opposition over privacy concerns. Position beacon messages are the main culprit, being broadcast in cleartext and pseudonymously signed up to 10 times per second. So far, no practical solutions have been proposed to encrypt or anonymously authenticate V2V messages. We propose two cryptographic innovations that enhance the privacy of V2V communication. As a core contribution, we introduce zone-encryption schemes, where vehicles generate and authentically distribute encryption keys associated to static geographic zones close to their location. Zone encryption provides security against eavesdropping, and, combined with a suitable anonymous authentication scheme, ensures that messages can only be sent by genuine vehicles, while adding only 224 Bytes of cryptographic overhead to each message. Our second contribution is an authentication mechanism fine-tuned to the needs of V2V which allows vehicles to authentically distribute keys, and is called dynamic group signatures with attributes. Our instantiation features unlimited locally generated pseudonyms, negligible credential download-and-storage costs, identity recovery by a trusted authority, and compact signatures of 216 Bytes at a 128-bit security level.

Abstract Pseudonymization is a widely deployed technique to de-sensitize data sets by consistently replacing identifying attributes with non-sensitive surrogates. However, all existing solutions are impractical to deploy in settings where data is accumulated from distributed sources: they either require sharing the same secret key with all sources, or rely on a fully trusted service to consistently compute these pseudonyms. Further, the consistency of pseudonyms, which is required to maintain the data’s utility, comes with inherent and severe privacy limitations. This paper solves the key management and privacy challenges by introducing oblivious pseudonymization-as-a-service. Therein, the pseudonymization is outsourced to a central, yet fully oblivious entity, i.e., the service neither learns the sensitive information nor the pseudonyms it produces. Further, to obtain better privacy we no longer require pseudonyms to be computed consistently and instead introduce a dedicated join procedure. When data is stored at rest, all data is pseudonymized in a fully unlinkable manner. Only when certain subsets of the data are needed, the linkage is established through a controlled and nontransitive join operation. We formally define the desired security properties in the UC framework and propose a generic protocol that provably satisfies them. The core of our scheme is a 3-party oblivious and convertible PRF, which we believe to be of independent interest.

The universal composability (UC) framework enables the modular design of cryptographic protocols by allowing arbitrary compositions of lower-level building blocks. Public-key encryption is unarguably a very important such building block. However, so far no UC-functionality exists that offers non-interactive encryption necessary for modular protocol construction. We provide an ideal functionality for non-committing encryption (i.e., public-key encryption secure against adaptive corruptions) with locally generated, and therefore non-interactive, ciphertexts. As a sanity check, we also provide a property-based security notion that we prove to be equivalent to the UC notion. We then show that the encryption scheme of Camenisch et al. (SCN '16) based on trapdoor permutations securely implements our notion in the random-oracle model without assuming secure erasures. This is the best one can hope to achieve as standard-model constructions do not exist due to the uninstantiability of round-optimal adaptively secure message transfer in the standard model (Nielsen, Crypto '02). We illustrate the modular reusability of our functionality by constructing the first non-interactive signcryption scheme secure against adaptive corruptions without secure erasures in the UC framework.

Personal information is often gathered and processed in a decentralized fashion. Examples include health records and governmental data bases. To protect the privacy of individuals, no unique user identifier should be used across the different databases. At the same time, the utility of the distributed information needs to be preserved which requires that it be nevertheless possible to link different records if they relate to the same user. Recently, Camenisch and Lehmann (CCS 15) have proposed a pseudonym scheme that addresses this problem by domain-specific pseudonyms. Although being unlinkable, these pseudonyms can be converted by a central authority (the converter). To protect the users' privacy, conversions are done blindly without the converter learning the pseudonyms or the identity of the user. Unfortunately, their scheme sacrifices a crucial privacy feature: transparency. Users are no longer able to inquire with the converter and audit the flow of their personal data. Indeed, such auditability appears to be diametral to the goal of blind pseudonym conversion. In this paper we address these seemingly conflicting requirements and provide a system where user-centric audits logs are created by the oblivious converter while maintaining all privacy properties. We prove our protocol to be UC-secure and give an efficient instantiation using novel building blocks.

The Trusted Platform Module (TPM) is an international standard for a security chip that can be used for the management of cryptographic keys and for remote attestation. The specification of the most recent TPM 2.0 interfaces for direct anonymous attestation unfortunately has a number of severe shortcomings. First of all, they do not allow for security proofs (indeed, the published proofs are incorrect). Second, they provide a Diffie-Hellman oracle w.r.t. the secret key of the TPM, weakening the security and preventing forward anonymity of attestations. Fixes to these problems have been proposed, but they create new issues: they enable a fraudulent TPM to encode information into an attestation signature, which could be used to break anonymity or to leak the secret key. Furthermore, all proposed ways to remove the Diffie-Hellman oracle either strongly limit the functionality of the TPM or would require significant changes to the TPM 2.0 interfaces. In this paper we provide a better specification of the TPM 2.0 interfaces that addresses these problems and requires only minimal changes to the current TPM 2.0 commands. We then show how to use the revised interfaces to build q-SDH-and LRSW-based anonymous attestation schemes, and prove their security. We finally discuss how to obtain other schemes addressing different use cases such as key-binding for U-Prove and e-cash.

When data maintained in a decentralized fashion needs to be synchronized or exchanged between different databases, related data sets usually get associated with a unique identifier. While this approach facilitates cross-domain data exchange, it also comes with inherent drawbacks in terms of controllability. As data records can easily be linked, no central authority can limit or control the information flow. Worse, when records contain sensitive personal data, as is for instance the case in national social security systems, such linkability poses a massive security and privacy threat. An alternative approach is to use domain-specific pseudonyms, where only a central authority knows the cross-domain relation between the pseudonyms. However, current solutions require the central authority to be a fully trusted party, as otherwise it can provide false conversions and exploit the data it learns from the requests. We propose an (un)linkable pseudonym system that overcomes those limitations, and enables controlled yet privacy-friendly exchange of distributed data. We prove our protocol secure in the UC framework and provide an efficient instantiation based on discrete-logarithm related assumptions.

We present a highly efficient cryptographic protocol to protect user passwords against server compromise by distributing the capability to verify passwords over multiple servers. Password verification is a single-round protocol and requires from each server only one exponentiation in a prime-order group. In spite of its simplicity, our scheme boasts security against dynamic and transient corruptions, meaning that servers can be corrupted at any time and can recover from corruption by going through a non-interactive key refresh procedure. The users' passwords remain secure against offline dictionary attacks as long as not all servers are corrupted within the same time period between refreshes. The only currently known scheme to achieve such strong security guarantees incurs the considerable cost of several hundred exponentiations per server. We prove our scheme secure in the universal composability model, which is well-known to offer important benefits for password-based primitives, under the gap one-more Diffie-Hellman assumption in the random-oracle model. Server initialization and refresh must take place in a trusted execution environment. Initialization additionally requires a secure message to each server, but the refresh procedure is non-interactive. We show that these requirements are easily met in practice by providing an example deployment architecture.