International Association
for Cryptologic Research

The goal of white-box cryptography is to protect secret keys embedded in a cryptographic software deployed in an untrusted environment. In this article, we revisit state-of-the-art countermeasures employed in white-box cryptography, and we discuss possible ways to combine them. Then we analyze the different gray-box attack paths and study their performances in terms of required traces and computation time. Afterward, we propose a new paradigm for the gray-box attack against white-box cryptography, which exploits the data-dependency of the target implementation. We demonstrate that our approach provides substantial complexity improvements over the existing attacks. Finally, we showcase this new technique by breaking the three winning AES-128 white-box implementations from WhibOx 2019 white-box cryptography competition.

The GIFT family of lightweight block ciphers, published at CHES 2017, offers excellent hardware performance figures and has been used, in full or in part, in several candidates of the ongoing NIST lightweight cryptography competition. However, implementation of GIFT in software seems complex and not efficient due to the bit permutation composing its linear layer (a feature shared with PRESENT cipher). In this article, we exhibit a new non-trivial representation of the GIFT family of block ciphers over several rounds. This new representation, that we call fixslicing, allows extremely efficient software bitsliced implementations of GIFT, using only a few rotations, surprisingly placing GIFT as a very efficient candidate on micro-controllers. Our constant time implementations show that, on ARM Cortex-M3, 128-bit data can be ciphered with only about 800 cycles for GIFT-64 and about 1300 cycles for GIFT-128 (assuming pre-computed round keys). In particular, this is much faster than the impressive PRESENT implementation published at CHES 2017 that requires 2116 cycles in the same setting, or the current best AES constant time implementation reported that requires 1617 cycles. This work impacts GIFT, but also improves software implementations of all other cryptographic primitives directly based on it or strongly related to it. For example, we observe that with the fixslicing technique GIFT-COFB performs faster than Ascon on ARM Cortex-M processors.

Secure multi-party computation has been extensively studied in the past years and has reached a level that is considered practical for several applications. The techniques developed thus far have been steadily optimized for performance and were shown to be secure in the classical setting, but are not known to be secure against quantum adversaries. In this work, we start to pave the way for secure two-party computation in a quantum world where the adversary has access to a quantum computer. We show that post-quantum secure two-party computation has comparable efficiency to their classical counterparts. For this, we develop a lattice-based OT protocol which we use to implement a post-quantum secure variant of Yao's famous garbled circuits (GC) protocol (FOCS'82). Along with the OT protocol, we show that the oblivious transfer extension protocol of Ishai et al. (CRYPTO'03), which allows running many OTs using mainly symmetric cryptography, is post-quantum secure. To support these results, we prove that Yao's GC protocol achieves post-quantum security if the underlying building blocks do.

We present the first practical software implementation of Supersingular Isogeny Key Encapsulation (SIKE) round 2, targeting NIST's 1, 2, and 5 security levels on 32-bit ARM Cortex-M4 microcontrollers. The proposed library introduces a new speed record of SIKE protocol on the target platform. We achieved this record by adopting several state-of-the-art engineering techniques as well as highly-optimized hand-crafted assembly implementation of finite field arithmetic. In particular, we carefully redesign the previous optimized implementations of filed arithmetic on 32-bit ARM Cortex-M4 platform and propose a set of novel techniques which are explicitly suitable for SIKE/SIDH primes. Moreover, the proposed arithmetic implementations are fully scalable to larger bit-length integers and can be adopted over different security levels. The benchmark result on STM32F4 Discovery board equipped with 32-bit ARM Cortex-M4 microcontrollers shows that the entire key encapsulation over p434 takes about 326 million clock cycles (i.e. 1.94 seconds @168MHz). In contrast to the previous optimized implementation of the isogeny-based key exchange on low-power 32-bit ARM Cortex-M4, our performance evaluation shows feasibility of using SIKE mechanism on the target platform. In comparison to the most of the post-quantum candidates, SIKE requires an excessive number of arithmetic operations, resulting in significantly slower timings. However, its small key size makes this scheme as a promising candidate on low-end microcontrollers in the quantum era by ensuring the lower energy consumption for key transmission than other schemes.

The US National Institute of Standards and Technology (NIST) recently announced the public-key cryptosystems (PKC) that have passed to the second round of the post-quantum standardization process. Most of these PKC come in two flavours: a weak IND-CPA version and a strongly secure IND-CCA construction. For the weaker scheme, no level of security is claimed in the plaintext-checking attack (PCA) model. However, previous works showed that, for several NIST candidates, only a few PCA queries are sufficient to recover the secret key. In order to create a more complete picture, we design new key-recovery PCA against several round 2 candidates. Our attacks against CRYSTALS-Kyber, HQC, LAC and SABER are all practical and require only a few thousand queries to recover the full secret key. In addition, we present another KR-PCA attack against the rank-based scheme RQC, which needs roughly $O(2^{38})$ queries. Hence, this type of scheme seems to resist better than others to key recovery. Motivated by this observation, we prove an interesting result on the rank metric. Namely, that the learning problem with the rank distance is hard for some parameters, thus invalidating a common strategy for reaction attacks.

Rainbow is a signature scheme that is based on multivariate polynomials. It is one of the Round-2 candidates of the NIST’s Post-Quantum Cryptography Standardization project. Its computations rely heavily on GF(2^8) arithmetic and the Rainbow submission optimizes the code by using AVX2 shuffle and permute instructions. In this paper, we show a new optimization that leverages: a) AVX512 architecture; b) the latest processor capabilities Galois Field New Instructions(GF-NI), available on Intel "Ice Lake" processor. We achieved a speedup of 2.43x/3.13x/0.64x for key generation/signing/verifying, respectively. We also propose a variation of Rainbow, with equivalent security, using a different representation of GF(2^8). With this variant, we achieve a speedup of 2.44x/4.7x/2.1x for key generation/signing/verifying, respectively.

With the growth of cloud computing, the need arises for Private Set Intersection (PSI) protocols that can operate on outsourced data and delegate computation to cloud servers. One limitation of existing delegated PSI protocols is that they are all designed for static data and do not allow efficient update on outsourced data. Another limitation is that they cannot efficiently support PSI among multiple clients, which is often needed in practice. This paper presents “Feather”, the first delegated PSI protocol that supports efficient data updates and scalable multi-party PSI computation on outsourced datasets. The clients can independently prepare and upload their private data to the cloud once, then delegate the computation an unlimited number of times. The update operation has O(1) communication and computation complexity, and this is achieved without sacrificing PSI efficiency and security. Feather does not use public key cryptography, that makes it more scalable. We have implemented a prototype and compared the concrete performance against the state of the art. The evaluation indicates that Feather does achieve better performance in both update and PSI computation.

The breakthrough of blockchain in many decentralized cryptocurrencies has reactivated studies on consensus under network synchrony, which has better security than a consensus under network asynchrony but has been considered to be impractical and only theoretical so far. One of the biggest concerns is the speed of transaction processing. To solve this concern, transactions can be processed responsively, i.e., without reliance on synchrony. Another approach is to minimize reliance on synchrony to achieve optimal synchronous latency. In this paper, we consider answering the question ”Can we achieve both responsiveness and optimal synchronous latency?”. To do this, we first show some theoretical possibilities and impossibilities in achieving both responsiveness and optimal synchronous latency, and then we present a practical blockchain or state machine replication protocol we call Hybrid-BFT. Hybrid-BFT can process transactions responsively under normal situation, i.e., small number of faults, while it can achieve optimal synchronous latency even under worst situation. Furthermore, Hybrid-BFT achieves responsive leader change, making it completely free from synchronous delay under normal situation.

Modern electronic voting systems (e-voting systems) are designed to provide not only vote privacy but also (end-to-end) verifiability. Several verifiable e-voting systems have been proposed in the literature, with Helios being one of the most prominent ones.

Almost all such systems, however, reveal not just the voting result but also the full tally, consisting of the exact number of votes per candidate or even all single votes. There are several situations where this is undesirable. For example, in elections with only a few voters (e.g., boardroom or jury votings), revealing the complete tally leads to a low privacy level, possibly deterring voters from voting for their actual preference. In other cases, revealing the complete tally might unnecessarily embarrass some candidates. Often, the voting result merely consists of a single winner or a ranking of candidates, so revealing only this information but not the complete tally is sufficient. This property is called tally-hiding and it offers completely new options for e-voting.

In this paper, we propose the first provably secure end-to-end verifiable tally-hiding e-voting system, called Ordinos. We instantiated our system with suitable cryptographic primitives, including an MPC protocol for greater-than tests, implemented the system, and evaluated its performance, demonstrating its practicality. Moreover, our work provides a deeper understanding of tally-hiding in general, in particular in how far tally-hiding affects the levels of privacy and verifiability of e-voting systems.

In this work we develop a rewarding framework that can be used as a building block in crowd-sensing applications. Although a core requirement of such systems is user engagement, people may be reluctant to participate as sensitive information about them may be leaked or inferred from submitted data. Thus monetary incentives could help attract a large number of participants, thereby increasing not only the amount but also the quality of sensed data. Our first contribution in this work is to ensure that users can submit data and obtain Bitcoin payments in a privacy-preserving manner, preventing curious providers from linking the data or the payments back to the user. At the same time, we thwart malicious user behavior such as double-redeeming attempts where a user tries to obtain rewards for multiple submissions of the same data. More importantly, we ensure the fairness of the exchange in a completely trustless manner; by relying on the Blockchain, we eliminate the trust placed on third parties in traditional fair exchange protocols. Finally, our system is highly efficient as most of the protocol steps do not utilize the Blockchain network. When they do, we only rely on simple Bitcoin transactions as opposed to prior works that are based on the use of highly complex smart contracts.

Chameleon-hash functions, introduced by Krawczyk and Rabin at NDSS 2000, are trapdoor collision-resistant hash-functions parametrized by a public key. If the corresponding secret key is known, arbitrary collisions for the hash function can be efficiently found. Chameleon-hash functions have prominent applications in the design of cryptographic primitives, such as lifting non-adaptively secure signatures to adaptively secure ones. Recently, this primitive also received a lot of attention as a building block in more complex cryptographic applications ranging from editable blockchains to advanced signature and encryption schemes.

We observe that in latter applications various different notions of collision-resistance are used, and it is not always clear if the respective notion does really cover what seems intuitively required by the application. Therefore, we revisit existing collision-resistance notions in the literature, study their relations, and - using the example of the recent redactable blockchain proposals - discuss which practical impact different notions of collision-resistance might have. Moreover, we provide a stronger, and arguably more desirable, notion of collision-resistance than what is known from the literature. Finally, we present a surprisingly simple and efficient black-box construction of chameleon-hash functions achieving this strong notion.

As part of our planned expansion in the area of Cybersecurity, the Department of Mathematics and Computer Science at the University of Southern Denmark, Odense, invites applications for an assistant professor position in Computer Science. We interpret the area cybersecurity broadly: the ideal applicant is a computer scientist with a strong theoretical background, conducting research on cybersecurity or research that has cybersecurity as a direct application. We would like to expand on our current security research efforts within Cryptology, Formal Methods, Information Security, Security in DevOps, and Data-driven approaches, so these topics are of particular interest. Application deadline: 30 May 2020.

Closing date for applications:

Contact: Jacopo Mauro mauro@imada.sdu.dk or Joan Boyar joan@imada.sdu.dk

More information: https://www.sdu.dk/da/service/ledige_stillinger/1098235

In this short note, we show that substantially weaker Low Order assumptions are sufficient to prove the soundness of Pietrzak's protocol for proof of exponentiation in groups of unknown order. This constitutes the first step to a better understanding of the asymptotic computational complexity of breaking the soundness of the protocol. Furthermore, we prove the equivalence of the (weaker) Low Order assumption(s) and the Factoring assumption in RSA groups for a non-negligible portion of moduli. We argue that in practice our reduction applies for a considerable amount of deployed moduli. Our results have cryptographic applications, most importantly in the theory of recently proposed verifiable delay function constructions. Finally, we describe how to certify RSA moduli free of low order elements.

Non-interactive zero-knowledge proofs, and more specifically succinct non-interactive zero-knowledge arguments (zk-SNARKS), have been proven to be the “swiss army knife” of the blockchain and distributed ledger space, with a variety of applications in privacy, interoperability and scalability. Many commonly used SNARK systems rely on a structured reference string, the secure generation of which turns out to be their Achilles heel: If the randomness used for the generation is known, the soundness of the proof system can be broken with devastating consequences for the underlying blockchain system that utilises them. In this work we describe and analyze, for the first time, a blockchain mechanism that produces a secure SRS with the characteristic that security is shown for the exact same conditions under which the blockchain protocol is proven to be secure. Our mechanism makes use of the recent discovery of updateable structure reference strings to perform this secure generation in a fully distributed manner. In this way, the SRS emanates from the normal operation of the blockchain protocol itself without the need of additional security assumptions or off-chain computation and/or verification. We provide concrete guidelines for the parameterisation of this system which allows for the completion of a secure setup in a reasonable period of time. We also provide an incentive scheme that, when paired with the update mechanism, properly incentivises participants into contributing to secure reference string generation.

In an article presented at FDTC 2018, Arribas, De Cnudde, and Sijacic prove under mild conditions that threshold implementations (TIs) are secure against fault sensitivity analysis (FSA). Later in 2018, in the PhD thesis of De Cnudde, additional assumptions were imposed to provably withstand FSA, thereby increasing the required number of random bits. We point out that even under the latter, stronger conditions, the proof remains incorrect.

To help fighting the COVID-19 pandemic, the Pan-European Privacy-Preserving Proximity Tracing (PEPP-PT) project proposed a Decentralized Privacy-Preserving Proximity Tracing (DP3T) system. This helps tracking the spread of SARS-CoV-2 virus while keeping the privacy of individuals safe. In this report, we analyze the security and the privacy protection of DP3T. Without questioning how effective it could be against the pandemic, we show that it may introduce severe risks to society. Furthermore, we argue that some privacy protection measurements by DP3T may have the opposite affect of what they were intended to. Specifically, sick and reported people may be deanonymized, private encounters may be revealed, and people may be coerced to reveal the private data they collect.

Contact tracing is a promising approach to combat the COVID -19 pandemic. Various systems have been proposed to automatise the process. However, user privacy was not a major design goal in most of these systems. Other designs rely heavily on a centralised server or reveal significant amounts of private data to health authorities. We propose CAUDHT, a decentralized peer-to-peer system for contact tracing. The central health authority can focus on providing and operating tests for the disease while contact tracing is done by the system’s users themselves. We use a distributed hash table to build a decentral messaging system for infected patients and their contacts. With blind signatures, we ensure that messages about infections are authentic and unchanged. A strong privacy focus enables data integrity, confidentiality, and privacy.

In this paper, we present all 4-bit S-boxes which are able to support BOGI logic. We exhaustively show that only 2,413 PXE classes of 4-bit S-box are BOGI-applicable among the 142,090,700 PXE classes. We evaluate the whole BOGI-applicable S-boxes in terms of the security and implementation costs. The security evaluation includes security strength of the S-boxes themselves, and how they affect the resistance of GIFT-64 against differential and linear cryptanalysis (DC and LC). The security evaluation shows that all the BOGI-applicable S-boxes fulfill the security criteria of GIFT designers as long as they have the differential uniformity and linearity as 6 and 8, respectively. It will also be shown that the security of GIFT-64 against DC and LC can be improved only by changing the S-box. Moreover, we evaluate the implementation costs of the BOGI-applicable S-boxes by finding their optimal implementation. The results show that GIFT S-box is well-chosen considering existence of fixed-points, and suggest a set of S-boxes providing the same implementation cost as GIFT S-box. Finally, we suggest a set of potentially better S-boxes for GIFT-64 based on our investigations.

In recent years, deep learning-based side-channel attacks have established its position as mainstream. However, most deep learning techniques for cryptanalysis mainly focused on classifying side-channel information in a profiled scenario where attackers can obtain the label of training data. In this paper, we introduce a novel approach with deep learning for improving side-channel attacks, especially in a non-profiling scenario. We also propose a new method that trains autoencoder through the noise from real data using the noise-reduced-label. It notably diminishes the noise in a trace by adapting the autoencoder framework to the signal preprocessing. We show the convincing comparison from our custom dataset, which captured that our works outperform conventional preprocessing methods such as principal component analysis and linear discriminant analysis. Furthermore, we apply the method to realign desynchronized traces that applied hiding countermeasures, and we experimentally validate the performance of the proposal. Also, for masking countermeasures, we experimentally show that we can improve the performance of side-channel analysis by using an existing combining function or proposed method using domain knowledge.

Reductions between problems, the mainstay of theoretical computer science, efficiently map an instance of one problem to an instance of another in such a way that solving the latter allows solving the former. The subject of this work is ``lossy'' reductions, where the reduction loses some information about the input instance. We show that such reductions, when they exist, have interesting and powerful consequences for lifting hardness into ``useful'' hardness, namely cryptography.

Our first, conceptual, contribution is a definition of lossy reductions in the language of mutual information. Roughly speaking, our definition says that a reduction $\mathsf{C}$ is $t$-lossy if, for any distribution $X$ over its inputs, the mutual information $I(X;\mathsf{C}(X)) \leq t$. Our treatment generalizes a variety of seemingly related but distinct notions such as worst-case to average-case reductions, randomized encodings (Ishai and Kushilevitz, FOCS 2000), homomorphic computations (Gentry, STOC 2009), and instance compression (Harnik and Naor, FOCS 2006).

We then proceed to show several consequences of lossy reductions:

1. We say that a language $L$ has an $f$-reduction to a language $L'$ for a Boolean function $f$ if there is a (randomized) polynomial-time algorithm $\mathsf{C}$ that takes an $m$-tuple of strings $X = (x_1,\ldots,x_m)$, with each $x_i\in\{0,1\}^n$, and outputs a string $z$ such that with high probability, \begin{align*} L'(z) = f(L(x_1),L(x_2),\ldots,L(x_m)) \end{align*} 2. Suppose a language $L$ has an $f$-reduction $\mathsf{C}$ to $L'$ that is $t$-lossy. Our first result is that one-way functions exist if $L$ is worst-case hard and one of the following conditions holds: - $f$ is the OR function, $t \leq m/100$, and $L'$ is the same as $L$ - $f$ is the Majority function, and $t \leq m/100$ - $f$ is the OR function, $t \leq O(m\log{n})$, and the reduction has no error

This improves on the implications that follow from combining (Drucker, FOCS 2012) with (Ostrovsky and Wigderson, ISTCS 1993) that result in {\em auxiliary-input} one-way functions.

3. Our second result is about the stronger notion of $t$-compressing $f$-reductions -- reductions that only output $t$ bits. We show that if there is an average-case hard language $L$ that has a $t$-compressing Majority reduction to some language for $t=m/100$, then there exist collision-resistant hash functions.

This improves on the result of (Harnik and Naor, STOC 2006), whose starting point is a cryptographic primitive (namely, one-way functions) rather than average-case hardness, and whose assumption is a compressing OR-reduction of SAT (which is now known to be false unless the polynomial hierarchy collapses).

Along the way, we define a non-standard one-sided notion of average-case hardness, which is the notion of hardness used in the second result above, that may be of independent interest.