Abstract Competitions are widely viewed as the safest way to select cryptographic algorithms. This paper surveys procedures that have been used in cryptographic competitions, and analyzes the extent to which those procedures reduce security risks.

Recent research has revealed an ever-growing class of microarchitectural attacks that exploit speculative execution, a standard feature in modern processors. Proposed and deployed countermeasures involve a variety of compiler updates, firmware updates, and hardware updates. None of the deployed countermeasures have convincing security arguments, and many of them have already been broken.

We introduce SPHINCS+, a stateless hash-based signature framework. SPHINCS+ has significant advantages over the state of the art in terms of speed, signature size, and security, and is among the nine remaining signature schemes in the second round of the NIST PQC standardization project. One of our main contributions in this context is a new few-time signature scheme that we call FORS. Our second main contribution is the introduction of tweakable hash functions and a demonstration how they allow for a unified security analysis of hash-based signature schemes. We give a security reduction for SPHINCS+ using this abstraction and derive secure parameters in accordance with the resulting bound. Finally, we present speed results for our optimized implementation of SPHINCS+ and compare to SPHINCS-256, Gravity-SPHINCS, and Picnic.

This paper introduces a high-security post-quantum stateless hash-based signature scheme that signs hundreds of messages per second on a modern 4-core 3.5GHz Intel CPU. Signatures are 41 KB, public keys are 1 KB, and private keys are 1 KB. The signature scheme is designed to provide long-term $$2^{128}$$ security even against attackers equipped with quantum computers. Unlike most hash-based designs, this signature scheme is stateless, allowing it to be a drop-in replacement for current signature schemes.

This paper shows how to securely authenticate messages using just $$29$$ bit operations per authenticated bit, plus a constant overhead per message. The authenticator is a standard type of “universal” hash function providing information-theoretic security; what is new is computing this type of hash function at very high speed. At a lower level, this paper shows how to multiply two elements of a field of size $$2^{128}$$ using just $$9062 \approx 71\,\cdot \, 128$$ bit operations, and how to multiply two elements of a field of size $$2^{256}$$ using just $$22164 \approx 87\,\cdot \, 256$$ bit operations. This performance relies on a new representation of field elements and new FFT-based multiplication techniques. This paper’s constant-time software uses just 1.89 Core 2 cycles per byte to authenticate very long messages. On a Sandy Bridge it takes 1.43 cycles per byte, without using Intel’s PCLMULQDQ polynomial-multiplication hardware. This is much faster than the speed records for constant-time implementations of GHASH without PCLMULQDQ (over 10 cycles/byte), even faster than Intel’s best Sandy Bridge implementation of GHASH with PCLMULQDQ (1.79 cycles/byte), and almost as fast as state-of-the-art 128-bit prime-field MACs using Intel’s integer-multiplication hardware (around 1 cycle/byte).

Censorship-circumvention tools are in an arms race against censors. The censors study all traffic passing into and out of their controlled sphere, and try to disable censorship-circumvention tools without completely shutting down the Internet. Tools aim to shape their traffic patterns to match unblocked programs, so that simple traffic profiling cannot identify the tools within a reasonable number of traces; the censors respond by deploying firewalls with increasingly sophisticated deep-packet inspection. Cryptography hides patterns in user data but does not evade censorship if the censor can recognize patterns in the cryptography itself. In particular, elliptic-curve cryptography often transmits points on known elliptic curves, and those points are easily distinguishable from uniform random strings of bits.

MinimaLT is a new network protocol that provides ubiquitous encryption for maximal confidentiality, including protecting packet headers. MinimaLT provides server and user authentication, extensive Denial-of-Service protections, privacy-preserving IP mobility, and fast key erasure. We describe the protocol, demonstrate its performance relative to TLS and unencrypted TCP/IP, and analyze its protections, including its resilience against DoS attacks. By exploiting the properties of its cryptographic protections, MinimaLT is able to eliminate three way handshakes and thus create connections faster than unencrypted TCP/IP.

This paper presents extremely fast algorithms for code-based public-key cryptography, including full protection against timing attacks. For example, at a 2 128 security level, this paper achieves a reciprocal decryption throughput of just 60493 cycles (plus cipher cost etc.) on a single Ivy Bridge core. These algorithms rely on an additive FFT for fast root computation, a transposed additive FFT for fast syndrome computation, and a sorting network to avoid cache-timing attacks.

NEON is a vector instruction set included in a large fraction of new ARM-based tablets and smartphones. This paper shows that NEON supports high-security cryptography at surprisingly high speeds; normally data arrives at lower speeds, giving the CPU time to handle tasks other than cryptography. In particular, this paper explains how to use a single 800MHz Cortex A8 core to compute the existing NaCl suite of high-security cryptographic primitives at the following speeds: 5.60 cycles per byte (1.14 Gbps) to encrypt using a shared secret key, 2.30 cycles per byte (2.78 Gbps) to authenticate using a shared secret key, 527102 cycles (1517/second) to compute a shared secret key for a new public key, 624846 cycles (1280/second) to verify a signature, and 244655 cycles (3269/second) to sign a message. These speeds make no use of secret branches and no use of secret memory addresses.