Not ECDSA. Not Schnorr. Meet DahLIAS.
Combination signatures aren’t new. They’ve been round because the early 2000s. However constructing one that really works in Bitcoin’s safety mannequin, with Bitcoin’s elliptic curve, has by no means been confirmed. Builders speculated it may be doable. They shared hand-wavy sketches and stated, “possibly it’d work like MuSig2, however throughout transaction inputs.” The concept lingered for years as developer folklore, shut, by no means provably confirmed.
That modified just lately, when Jonas Nick and Tim Ruffing of Blockstream Analysis, along with Yannick Seurin of Ledger, revealed a paper that turned this cryptographic ghost story right into a concrete, provable end result. DahLIAS is the primary formal, safe building of a full constant-size combination signature (CISA) scheme that works on Bitcoin’s native curve!
However that’s a variety of phrases, so let’s break that down:
- Full aggregation: A number of signatures throughout totally different inputs are mixed into one — and the result’s a 64 byte signature whose measurement stays fixed, irrespective of what number of signers or inputs.
- Cross-input: Every signer can authorize totally different inputs, and all mix into one signature.
It provides no vital new assumptions past these already relied on by Bitcoin. DahLIAS builds a brand new cryptographic primitive utilizing the identical math Bitcoin already depends on, unlocking a wholly new type of signature.
Let’s Discuss About Curves and Signatures
Digital signatures are how Bitcoin proves {that a} consumer has licensed a transaction. While you go to spend bitcoin, your pockets makes use of a non-public key to signal a message, and the community verifies that signature utilizing the matching public key.
Bitcoin makes use of the secp256k1 curve. It’s quick, environment friendly, and has been battle-tested over time. It helps signature schemes like ECDSA (Bitcoin’s unique signature algorithm) and Schnorr (added by Taproot in 2021), that are at the moment the one signature schemes permitted by Bitcoin consensus.
Historically, full signature aggregation relied on mathematical operations not supported by Bitcoin’s curve, secp256k1, which made it appear out of attain. These options have sometimes relied on different kinds of elliptic curves. For instance, BLS (Boneh–Lynn–Shacham) signatures use a particular type of curve referred to as a pairing-friendly curve, which permits superior operations like combining many signatures, even on totally different messages, into one.
The issue is that BLS signatures don’t work on secp256k1. Whereas Schnorr was a pure improve from ECDSA, since each depend on the identical type of elliptic curve, including BLS could be a a lot greater leap and a departure from Bitcoin’s current safety mannequin. Although technically doable, it could introduce new cryptographic assumptions and add vital complexity to the protocol. Supporting a curve that’s pairing-friendly, like BLS12-381, could be a significant change for Bitcoin.
That is a part of why full signature aggregation has by no means been executed on secp256k1.
Till now.
What Combination Signatures Really Do
Most Bitcoin customers are accustomed to multisignatures. In a multisig pockets, a number of individuals collectively authorize the spending of a single UTXO or some particular “coin”. Everybody indicators the identical enter information. This setup is beneficial for issues like shared custody wallets.
Combination signatures work in a different way. As an alternative of a number of individuals signing the identical enter or coin, every signer authorizes a special UTXO in a transaction. These separate signatures are then compressed into one compact proof. With DahLIAS, meaning a single 64-byte signature on Bitcoin’s secp256k1 curve that verifies all inputs directly.
Meaning if in case you have 5 inputs from 5 totally different individuals, the transaction wants 5 totally different signatures. With an combination signature, all of these could be bundled into one. Even when every signer is spending a special enter and signing a special a part of the transaction, the result’s one signature that proves your complete transaction was correctly licensed.
It’s like zipping a complete record of approvals into one file. The signature is compact, however nonetheless verifiably proves that every signer licensed their particular UTXO.
As an alternative of verifying 10 separate signatures, you confirm one.
This helps realign incentives for privateness. By lowering the signature overhead to a single 64-byte proof, DahLIAS lowers the price of combining inputs in CoinJoins, making it financially smarter to decide on privateness than to go with out it.
Why Half-Aggregation Received Shut
Shortly after Schnorr signatures had been launched on Bitcoin, builders explored half-aggregation, as a option to compress a number of signatures however they weren’t fastened measurement. Every enter contributes to the dimensions of the signature, so the transaction nonetheless grows with each participant. DahLIAS fixes this by enabling full-aggregation throughout inputs and signers. Irrespective of how many individuals are concerned or what they’re signing, all their signatures compress into one constant-size, 64-byte proof.
What DahLIAS Really Unlocks
The principle profit right here is that DahLIAS are lowering the dimensions of advanced transactions.
DahLIAS makes use of a two-round interactive signing course of. It’s just like MuSig2 in that regard, nevertheless it isn’t a multisignature protocol as a result of it doesn’t require all members to co-sign the identical message. As an alternative, it aggregates totally different signatures on totally different messages throughout the transaction.
DahLIAS can also be sooner to confirm than checking every signature individually, as much as twice as quick in some instances. Decrease verification prices make it simpler for extra individuals to run full nodes, which helps protect Bitcoin’s decentralization over time.
Importantly, DahLIAS comes with robust cryptographic ensures. The scheme consists of formal safety proofs. Earlier ‘folklore’ approaches to full signature aggregation lacked this, and a few had been even later proven to be insecure. Fortuitously they weren’t adopted prematurely.
It’s value repeating: DahLIAS is just not a multisig protocol. It isn’t similar to MuSig2 or FROST from a purposeful standpoint, even when it shares comparable cryptographic constructing blocks. It serves a special function. It gives a brand new option to encode many impartial approvals into one clear, verifiable bundle.
Future Instructions
You may assume: if DahLIAS is so highly effective, why isn’t it a BIP? Why not suggest it for Bitcoin consensus?
DahLIAS signatures don’t appear like Schnorr or ECDSA signatures. The verification algorithm is totally different. As an alternative of taking a single public key, message, and signature, a DahLIAS verifier takes lists of public keys and messages, and a single 64-byte proof.
This makes DahLIAS incompatible with Bitcoin’s present consensus guidelines. Supporting it on the base layer would require a consensus change. This paper doesn’t suggest that change, nevertheless it does one thing equally necessary.
This paper reveals {that a} full signature aggregation scheme for Bitcoin’s native curve is feasible.
That alone is a significant step ahead.
To make DahLIAS a part of Bitcoin, somebody would want to write down a Bitcoin Enchancment Proposal (BIP), possibly even utilizing secp256k1lab. Meaning specifying the scheme intimately, contemplating its implications for consensus and implementation, and constructing group assist. This paper lays the cryptographic basis for that dialog.
The true worth of the DahLIAS paper is what it proves. Full signature aggregation on secp256k1 is not only a thought experiment. It’s concrete. It’s environment friendly. It’s safe. For years, the concept lived in developer folklore. Now, it’s written down, analyzed, and confirmed. All that’s left is to carry it to Bitcoin—if we wish it.
This can be a visitor put up by Kiara Bickers. Opinions expressed are fully their very own and don’t essentially replicate these of BTC Inc or Bitcoin Journal.
This put up Not ECDSA. Not Schnorr. Meet DahLIAS. first appeared on Bitcoin Magazine and is written by Kiara Bickers.