International Association
for Cryptologic Research

The Peregrine signature scheme is one of the candidates in the ongoing Korean post-quantum cryptography competition. It is proposed as a high-speed variant of Falcon, which is a hash-and-sign signature scheme over NTRU lattices and one of the schemes selected by NIST for standardization. To this end, Peregrine replaces the lattice Gaussian sampler in the Falcon signing procedure with a new sampler based on the centered binomial distribution. While this modification offers significant advantages in terms of efficiency and implementation, it does not come with a provable guarantee that signatures do not leak information about the signing key. Unfortunately, lattice based signature schemes in the hash-and-sign paradigm that lack such a guarantee (such as GGH, NTRUSign or DRS) have generally proved insecure. In this paper, we show that Peregrine is no exception, by demonstrating a practical key recovery attack against it. We observe that the distribution of Peregrine signatures is a hidden transformation of some public distribution and still leaks information about the signing key. By adapting the parallelepiped-learning technique of Nguyen and Regev (Eurocrypt 2006), we show that the signing key can be recovered from a relatively small number of signatures. The learning technique alone yields an approximate version of the key, from which we can recover the exact key using a decoding technique due to Thomas Prest (PKC 2023). For the reference implementation (resp. the official specification version) of Peregrine-512, we fully recover the secret key with good probability in a few hours given around 25,000 (resp. 11 million) signature samples.

We introduce a toolkit for transforming lattice-based hash-and-sign signature schemes into masking-friendly signatures secure in the t-probing model. Until now, efficiently masking lattice-based hash-and-sign schemes was an open problem unsuccessful attempts such as Mitaka. Our toolkit includes noise flooding to mitigate statistical leaks and an extended Strong Non-Interfering probing security property (SNIu) for masking gadgets to handle unshared inputs. Our main conceptual contribution lies in finding a systematic way to use noise flooding within the hash-and-sign paradigm. Our main technical contribution is to formalize, prove, instantiate and implement a hash-and-sign scheme based on these techniques. We showcase the efficiency of our techniques in a signature scheme, Plover-RLWE, based on (hint) Ring-LWE. It is the first lattice-based masked hash-and-sign scheme with quasi-linear complexity O(d log d) in the number of shares d. Our performances are competitive with the state-of-the-art masking-friendly signature, the Fiat-Shamir scheme Raccoon.

Threshold signatures have recently seen a renewed interest due to applications in cryptocurrency while NIST has released a call for multi-party threshold schemes, with a deadline for submission expected for the first half of 2025. So far, all lattice-based threshold signatures requiring less than two-rounds are based on heavy tools such as (fully) homomorphic encryption (FHE) and homomorphic trapdoor commitments (HTDC). This is not unexpected considering that most efficient two-round signatures from classical assumptions either rely on idealized model such as algebraic group models or on one-more type assumptions, none of which we have a nice analogue in the lattice world. In this work, we construct the first efficient two-round lattice-based threshold signature without relying on FHE or HTDC. It has an offline- online feature where the first round can be reprocessed without knowing message or the signer sets, effectively making the signing phase non-interactive. The signature size is small and shows great scalability. For example, even for a threshold as large as 1024 signers, we achieve a signature size roughly 11 KB. At the heart of our construction is a new lattice-based assumption called the algebraic one-more learning with errors (AOM-MLWE) assumption. We believe this to be a strong inclusion to our lattice toolkits with an independent interest. We establish the selective security of AOM-MLWE based on the standard MLWE and MSIS assumptions, and provide an in depth analysis of its adaptive security, which our threshold signature is based on.

We propose a new framework based on random submersions — that is projection over a random subspace blinded by a small Gaussian noise — for constructing verifiable short secret sharing and showcase it to construct efficient threshold lattice-based signatures in the hash-and-sign paradigm, when based on noise flooding. This is, to our knowledge, the first hash-and-sign lattice-based threshold signature. Our threshold signature enjoys the very desirable property of robustness, including at key generation. In practice, we are able to construct a robust hash-and-sign threshold signature for threshold and provide a typical parameter set for threshold T = 16 and signature size 13kB. Our constructions are provably secure under standard MLWE assumption in the ROM and only require basic primitives as building blocks. In particular, we do not rely on FHE-type schemes.

We present cryptanalysis of the inhomogenous short integer solution (ISIS) problem for anomalously small moduli q by exploiting the geometry of BKZ reduced bases of q-ary lattices. We apply this cryptanalysis to examples from the literature where taking such small moduli has been suggested. A recent work [Espitau--Tibouchi--Wallet--Yu, CRYPTO 2022] suggests small q versions of the lattice signature scheme Falcon and its variant Mitaka. For one small q parametrisation of Falcon we reduce the estimated security against signature forgery by approximately 26 bits. For one small q parametrisation of Mitaka we successfully forge a signature in 15 seconds.

We present a general framework for polynomial-time lattice Gaussian sampling. It revolves around a systematic study of the discrete Gaussian measure and its samplers under \emph{extensions} of lattices; we first show that given lattices $\Lat'\subset \Lat$ we can sample efficiently in $\Lat$ if we know how to do so in $\Lat'$ and the quotient $\Lat/\Lat'$, \emph{regardless} of the primitivity of $\Lat'$. As a direct application, we tackle the problem of domain extension and restriction for sampling and propose a sampler tailored for lattice \emph{filtrations}, which can be seen as a broad generalization of the celebrated Klein's sampler. Then, we demonstrate how to sample using a change of bases, or even switching the ambient space, even when the target lattice is not represented as full-rank in the ambient space. We show how to correct the induced distortion with the ``convolution-like'' technique of Peikert (Crypto 2010) (which we encompass as a byproduct). Since our framework aims at modularity and leverage the combinations of smaller samplers to build new ones, we also propose ad-hoc samplers for the so-called \emph{root lattices} $\An_n, \Dn_n, \mathsf{E}_n$ as base cases, extending the state-of-the-art for root lattice sampling, which was limited to $\ZZ^n$. We also show how our framework blends with the so-called $k$ing construction and provides a sampler for the remarkable Leech and Barnes-Wall lattices. As a by-product, we obtain novel, quasi-linear samplers for prime and smooth conductor (as $2^\ell 3^k$) cyclotomic rings, achieving essentially optimal Gaussian width. In a practice-oriented application, we showcase the impact of our work on hash-and-sign signatures over \textsc{ntru} lattices. In the best case, we can gain around 200 bytes (which corresponds to an improvement greater than 20\%) on the signature size. We also improve the new gadget-based constructions (Yu, Jia, Wang, Crypto 2023) and gain up to 110 bytes for the resulting signatures. Lastly, we sprinkle our exposition with several new estimates for the smoothing parameter of lattices, stemming from our algorithmic constructions and by novel methods based on series reversion.

In this paper, we introduce a novel trapdoor generation technique for Prest's hybrid sampler over NTRU lattices. Prest's sampler is used in particular in the recently proposed Mitaka signature scheme (Eurocrypt 2022), a variant of the Falcon signature scheme, one of the candidates selected by NIST for standardization. Mitaka was introduced to address Falcon's main drawback, namely the fact that the lattice Gaussian sampler used in its signature generation is highly complex, difficult to implement correctly, to parallelize or protect against side-channels, and to instantiate over rings of dimension not a power of two to reach intermediate security levels. Prest's sampler is considerably simpler and solves these various issues, but when applying the same trapdoor generation approach as Falcon, the resulting signatures have far lower security in equal dimension. The Mitaka paper showed how certain randomness-recycling techniques could be used to mitigate this security loss, but the resulting scheme is still substantially less secure than Falcon (by around 20 to 50 bits of CoreSVP security depending on the parameters), and has much slower key generation. Our new trapdoor generation techniques solves all of those issues satisfactorily: it gives rise to a much simpler and faster key generation algorithm than Mitaka's (achieving similar speeds to Falcon), and is able to comfortably generate trapdoors reaching the same NIST security levels as Falcon as well. It can also be easily adapted to rings of intermediate dimensions, in order to support the same versatility as Mitaka in terms of parameter selection. All in all, this new technique combines all the advantages of both Falcon and Mitaka (and more) with none of the drawbacks.

Recently, numerous physical attacks have been demonstrated against lattice-based schemes, often exploiting their unique properties such as the reliance on Gaussian distributions, rejection sampling and FFT-based polynomial multiplication. As the call for concrete implementations and deployment of postquantum cryptography becomes more pressing, protecting against those attacks is an important problem. However, few countermeasures have been proposed so far. In particular, masking has been applied to the decryption procedure of some lattice-based encryption schemes, but the much more difficult case of signatures (which are highly nonlinear and typically involve randomness) has not been considered until now. In this paper, we describe the first masked implementation of a lattice-based signature scheme. Since masking Gaussian sampling and other procedures involving contrived probability distributions would be prohibitively inefficient, we focus on the GLP scheme of Güneysu, Lyubashevsky and Pöppelmann (CHES 2012). We show how to provably mask it in the Ishai–Sahai–Wagner model (CRYPTO 2003) at any order in a relatively efficient manner, using extensions of the techniques of Coron et al. for converting between arithmetic and Boolean masking. Our proof relies on a mild generalization of probing security that supports the notion of public outputs. We also provide a proof-of-concept implementation to assess the efficiency of the proposed countermeasure.

The lattice reduction attack on (EC)DSA (and other Schnorr-like signature schemes) with partially known nonces, originally due to Howgrave-Graham and Smart, has been at the core of many concrete cryptanalytic works, side-channel based or otherwise, in the past 20 years. The attack itself has seen limited development, however: improved analyses have been carried out, and the use of stronger lattice reduction algorithms has pushed the range of practically vulnerable parameters further, but the lattice construction based on the signatures and known nonce bits remain the same.In this paper, we propose a new idea to improve the attack based on the same data in exchange for additional computation: carry out an exhaustive search on some bits of the secret key. This turns the problem from a single bounded distance decoding (BDD) instance in a certain lattice to multiple BDD instances in a fixed lattice of larger volume but with the same bound (making the BDD problem substantially easier). Furthermore, the fact that the lattice is fixed lets us use batch/preprocessing variants of BDD solvers that are far more efficient than repeated lattice reductions on non-preprocessed lattices of the same size. As a result, our analysis suggests that our technique is competitive or outperforms the state of the art for parameter ranges corresponding to the limit of what is achievable using lattice attacks so far (around 2-bit leakage on 160-bit groups, or 3-bit leakage on 256-bit groups).We also show that variants of this idea can also be applied to bits of the nonces (leading to a similar improvement) or to filtering signature data (leading to a data-time trade-off for the lattice attack). Finally, we use our technique to obtain an improved exploitation of the TPM–FAIL dataset similar to what was achieved in the Minerva attack.

This work describes the Mitaka signature scheme: a new hash-and-sign signature scheme over NTRU lattices which can be seen as a variant of NIST finalist Falcon. It achieves comparable efficiency but is considerably simpler, online/offline, and easier to parallelize and protect against side-channels, thus offering significant advantages from an implementation standpoint. It is also much more versatile in terms of parameter selection. We obtain this signature scheme by replacing the FFO lattice Gaussian sampler in Falcon by the “hybrid” sampler of Ducas and Prest, for which we carry out a detailed and corrected security analysis. In principle, such a change can result in a substantial security loss, but we show that this loss can be largely mitigated using new techniques in key generation that allow us to construct much higher quality lattice trapdoors for the hybrid sampler relatively cheaply. This new approach can also be instantiated on a wide variety of base fields, in contrast with Falcon's restriction to power-of-two cyclotomics. We also introduce a new lattice Gaussian sampler with the same quality and efficiency, but which is moreover compatible with the integral matrix Gram root technique of Ducas et al., allowing us to avoid floating point arithmetic. This makes it possible to realize the same signature scheme as Mitaka efficiently on platforms with poor support for floating point numbers. Finally, we describe a provably secure masking of Mitaka. More precisely, we introduce novel gadgets that allow provable masking at any order at much lower cost than previous masking techniques for Gaussian sampling-based signature schemes, for cheap and dependable side-channel protection.

Lattice-based digital signature schemes following the hash-and-sign design paradigm of Gentry, Peikert and Vaikuntanathan (GPV) tend to offer an attractive level of efficiency, particularly when instantiated with structured compact trapdoors. In particular, NIST postquantum finalist Falcon is both quite fast for signing and verification and quite compact: NIST notes that it has the smallest bandwidth (as measured in combined size of public key and signature) of all round 2 digital signature candidates. Nevertheless, while Falcon--512, for instance, compares favorably to ECDSA--384 in terms of speed, its signatures are well over 10 times larger. For applications that store large number of signatures, or that require signatures to fit in prescribed packet sizes, this can be a critical limitation. In this paper, we explore several approaches to further improve the size of hash-and-sign lattice-based signatures, particularly instantiated over NTRU lattices like Falcon and its recent variant Mitaka. In particular, while GPV signatures are usually obtained by sampling lattice points according to some *spherical* discrete Gaussian distribution, we show that it can be beneficial to sample instead according to a suitably chosen *ellipsoidal* discrete Gaussian: this is because only half of the sampled Gaussian vector is actually output as the signature, while the other half is recovered during verification. Making the half that actually occurs in signatures shorter reduces signature size at essentially no security loss (in a suitable range of parameters). Similarly, we show that reducing the modulus $q$ with respect to which signatures are computed can improve signature size as well as verification key size almost ``for free''; this is particularly true for constructions like Falcon and Mitaka that do not make substantial use of NTT-based multiplication (and rely instead on transcendental FFT). Finally, we show that the Gaussian vectors in signatures can be represented in a more compact way with appropriate coding-theoretic techniques, improving signature size by an additional 7 to 14%. All in all, we manage to reduce the size of, e.g., Falcon signatures by 30--40% at the cost of only 4--6 bits of Core-SVP security.

The LLL algorithm is a polynomial-time algorithm for reducing d-dimensional lattice with exponential approximation factor. Currently, the most efficient variant of LLL, by Neumaier and Stehl\'e, has a theoretical running time in $d^4\cdot B^{1+o(1)}$ where $B$ is the bitlength of the entries, but has never been implemented. This work introduces new asymptotically fast, parallel, yet heuristic, reduction algorithms with their optimized implementations. Our algorithms are recursive and fully exploit fast block matrix multiplication. We experimentally demonstrate that by carefully controlling the floating-point precision during the recursion steps, we can reduce euclidean lattices of rank d in time $\tilde{O}(d^\omega\cdot C)$, i.e., almost a constant number of matrix multiplications, where $\omega$ is the exponent of matrix multiplication and C is the log of the condition number of the matrix. For cryptographic applications, C is close to B, while it can be up to d times larger in the worst case. It improves the running-time of the state-of-the-art implementation fplll by a multiplicative factor of order $d^2\cdot B$. Further, we show that we can reduce structured lattices, the so-called knapsack lattices, in time $\tilde{O}(d^{\omega-1}\cdot C)$ with a progressive reduction strategy. Besides allowing reducing huge lattices, our implementation can break several instances of Fully Homomorphic Encryption schemes based on large integers in dimension 2,230 with 4 millions of bits.

We introduce a framework generalizing lattice reduction algorithms to module lattices in order to \emph{practically} and \emph{efficiently} solve the $\gamma$-Hermite Module-SVP problem over arbitrary cyclotomic fields. The core idea is to exploit the structure of the subfields for designing a recursive strategy of reduction in the tower of fields we are working in. Besides, we demonstrate how to leverage the inherent symplectic geometry existing such fields to provide a significant speed-up of the reduction for rank two modules. As a byproduct, we also generalize to all cyclotomic fields and provide speedups for many previous number theoretical algorithms, in particular to the rounding in the so-called Log-unit lattice. Quantitatively, we show that a module of rank 2 over a cyclotomic field of degree $n$ can be heuristically reduced within approximation factor $2^{\tilde{O}(n)}$ in time $\tilde{O}(n^2B)$, where $B$ is the bitlength of the entries. For $B$ large enough, this complexity shrinks to $\tilde{O}(n^{\log_2 3}B)$. This last result is particularly striking as it goes below the estimate of $n^2B$ swaps given by the classical analysis of the LLL algorithm using the decrease of the \emph{potential} of the basis. Finally, all this framework is fully parallelizable, and we provide a full implementation. We apply it to break multilinear cryptographic candidates on concrete proposed parameters. We were able to reduce matrices of dimension 4096 with 6675-bit integers in 4 days, which is more than a million times faster than previous state-of-the-art implementations. Eventually, we demonstrate a quasicubic time for the Gentry-Szydlo algorithm which finds a generator given the relative norm and a basis of an ideal. This algorithm is important in cryptanalysis and requires efficient ideal multiplications and lattice reductions; as such we can practically use it in dimension 1024.

This paper is devoted to analyzing the variant of Regev’s learning with errors (LWE) problem in which modular reduction is omitted: namely, the problem (ILWE) of recovering a vector $$\mathbf {s}\in \mathbb {Z}^n$$ given polynomially many samples of the form $$(\mathbf {a},\langle \mathbf {a},\mathbf {s}\rangle + e)\in \mathbb {Z}^{n+1}$$ where $$\mathbf { a}$$ and e follow fixed distributions. Unsurprisingly, this problem is much easier than LWE: under mild conditions on the distributions, we show that the problem can be solved efficiently as long as the variance of e is not superpolynomially larger than that of $$\mathbf { a}$$. We also provide almost tight bounds on the number of samples needed to recover $$\mathbf {s}$$.Our interest in studying this problem stems from the side-channel attack against the BLISS lattice-based signature scheme described by Espitau et al. at CCS 2017. The attack targets a quadratic function of the secret that leaks in the rejection sampling step of BLISS. The same part of the algorithm also suffers from a linear leakage, but the authors claimed that this leakage could not be exploited due to signature compression: the linear system arising from it turns out to be noisy, and hence key recovery amounts to solving a high-dimensional problem analogous to LWE, which seemed infeasible. However, this noisy linear algebra problem does not involve any modular reduction: it is essentially an instance of ILWE, and can therefore be solved efficiently using our techniques. This allows us to obtain an improved side-channel attack on BLISS, which applies to 100% of secret keys (as opposed to $${\approx }7\%$$ in the CCS paper), and is also considerably faster.