Untraceable transactions (UTT)

tl;dr: UTT is Chaumiam ecash done the right way: efficiently, decentralized and with sensible-anonymity: a user can only send (say) $\$L$ coins per month (assuming a Sybil-resistant proof-of-humanity system). UTT is academic work that started in 2018 at VMware, was rebooted in 2021 and now lives as an eprint¹. UTT did see deployment in a testing pilot with Israel’s central bank²$^,$³. Below, I’ll post some links to some slides and talks. Hopefully this will become a full blog post later.

Talks & slides

• SBC’23 slides here
• Stanford Security Seminar slides here

Blogposts

UTT relies on many cryptographic building blocks. Here are relevant blog posts:

Cryptographic primitives at the core of UTT:

• Pedersen commitments — coins are represented as Pedersen commitments to (pid, sn, value)
• Blind and threshold signatures — Chaum’s blind signatures are the foundation of ecash
• Pointcheval-Sanders (PS) signatures — used as the rerandomizable signature scheme for signing coins
• Sigma protocols — used for proving correctness of nullifiers and authorizing spends
• Zero-knowledge proofs — the foundation for transaction privacy
• Anonymous credentials — UTT’s registration credentials are a form of anonymous credential
• Range proofs from polynomial commitments — an alternative approach to range proofs
• DeKART range proofs — used to prove value preservation (no money creation)
• Identity-based encryption (IBE) — used to encrypt coin values to recipients using user-friendly identifiers (e.g., phone numbers)
• ElGamal encryption — related encryption primitive
• Polynomial secret sharing — background on Shamir secret sharing used in threshold cryptography

Relevant blog posts from decentralizedthoughts.github.io:

• A simple zero-knowledge proof — introductory ZK proof concepts
• Simple anonymous payments — a simplified model of anonymous payment systems

Code

There is actually an old public implementation of UTT as a library, initially written by me, which was integrated in VMware’s Concord BFT engine: see https://github.com/vmware/concord-bft/tree/master/utt/libutt

References

For cited works, see below 👇👇