Starkad and Poseidon: New Hash Functions for Zero Knowledge Proof Systems

Research output: Working paperResearch

Abstract

The area of practical proof systems, like SNARKs, STARKs, or Bulletproofs, is seeing a very dynamic development. Many use-cases of such systems involve, often as their most expensive apart, proving the knowledge of a preimage under a certain cryptographic hash function.

In this paper we present a modular framework and concrete instances of cryptographic hash functions which either work natively with GF(p) objects or on binary strings. Compared to competitors, our hash function Poseidon uses up to 8x fewer constraints per message bit compared to Pedersen Hash, whereas our STARK-friendly hash Starkad takes wins the factor of 4 over the hash function Friday by using a much smaller field.
Original languageEnglish
Publication statusPublished - 6 May 2019

Fingerprint

Hash functions
Concretes

Keywords

  • snark
  • stark
  • bulletproofs
  • hash
  • zkp

Cite this

@techreport{06fa956a7e5c48a7aeaae3fda86eff09,
title = "Starkad and Poseidon: New Hash Functions for Zero Knowledge Proof Systems",
abstract = "The area of practical proof systems, like SNARKs, STARKs, or Bulletproofs, is seeing a very dynamic development. Many use-cases of such systems involve, often as their most expensive apart, proving the knowledge of a preimage under a certain cryptographic hash function.In this paper we present a modular framework and concrete instances of cryptographic hash functions which either work natively with GF(p) objects or on binary strings. Compared to competitors, our hash function Poseidon uses up to 8x fewer constraints per message bit compared to Pedersen Hash, whereas our STARK-friendly hash Starkad takes wins the factor of 4 over the hash function Friday by using a much smaller field.",
keywords = "snark, stark, bulletproofs, hash, zkp",
author = "Lorenzo Grassi and Daniel Kales and Dmitry Khovratovich and Arnab Roy and Christian Rechberger and Markus Schofnegger",
year = "2019",
month = "5",
day = "6",
language = "English",
type = "WorkingPaper",

}

TY - UNPB

T1 - Starkad and Poseidon: New Hash Functions for Zero Knowledge Proof Systems

AU - Grassi, Lorenzo

AU - Kales, Daniel

AU - Khovratovich, Dmitry

AU - Roy, Arnab

AU - Rechberger, Christian

AU - Schofnegger, Markus

PY - 2019/5/6

Y1 - 2019/5/6

N2 - The area of practical proof systems, like SNARKs, STARKs, or Bulletproofs, is seeing a very dynamic development. Many use-cases of such systems involve, often as their most expensive apart, proving the knowledge of a preimage under a certain cryptographic hash function.In this paper we present a modular framework and concrete instances of cryptographic hash functions which either work natively with GF(p) objects or on binary strings. Compared to competitors, our hash function Poseidon uses up to 8x fewer constraints per message bit compared to Pedersen Hash, whereas our STARK-friendly hash Starkad takes wins the factor of 4 over the hash function Friday by using a much smaller field.

AB - The area of practical proof systems, like SNARKs, STARKs, or Bulletproofs, is seeing a very dynamic development. Many use-cases of such systems involve, often as their most expensive apart, proving the knowledge of a preimage under a certain cryptographic hash function.In this paper we present a modular framework and concrete instances of cryptographic hash functions which either work natively with GF(p) objects or on binary strings. Compared to competitors, our hash function Poseidon uses up to 8x fewer constraints per message bit compared to Pedersen Hash, whereas our STARK-friendly hash Starkad takes wins the factor of 4 over the hash function Friday by using a much smaller field.

KW - snark

KW - stark

KW - bulletproofs

KW - hash

KW - zkp

M3 - Working paper

BT - Starkad and Poseidon: New Hash Functions for Zero Knowledge Proof Systems

ER -