# PQLN: Post-Quantum Security for the Bitcoin Lightning Network's Off-Chain Surfaces

Crypto.world

> A future quantum computer running Shor's algorithm would break the secp256k1 elliptic-curve cryptography that Lightning uses for node identity, transport encryption, invoices and per-hop onion keys…

**arXiv**: 2609.13781

**Published**: 2026-09-12

**Authors**: Ahmet Kurt, Abdul-Salem Beibitkhan, Yacoub Hanna, Abdullah Aydeger

## AI summary

## Overview

**Research area:** Post-quantum cryptography applied to Bitcoin's second layer, specifically the off-chain protocol surfaces of the Lightning Network (gossip, transport, invoices, offers, payment onions).

**Technical level:** Advanced. The paper assumes familiarity with elliptic-curve cryptography, Noise handshakes, Sphinx onion routing, and the BOLT specification series, though its prose explanations of each mechanism are accessible to a determined non-specialist.

**Scope:** The paper designs, formally argues the security of, implements in rust-lightning, and empirically evaluates PQLN, a hybrid post-quantum extension covering Lightning's off-chain surfaces while interoperating with unmodified nodes. (Note: the supplied text cuts off mid-way through Assumption 2 in Section V, so the full proof sketches are not present in the content reviewed here.)

## What This Paper Is About

A future quantum computer running Shor's algorithm would break the secp256k1 elliptic-curve cryptography that Lightning uses for node identity, transport encryption, invoices and per-hop onion keys, and an adversary can already record encrypted traffic today to decrypt it later. Because these mechanisms live purely off-chain, they cannot be fixed by a Bitcoin consensus change, and no post-quantum Lightning design, implementation or evaluation existed before this work. PQLN's goal is to add post-quantum protection to all of these off-chain surfaces now, using the NIST standards ML-DSA and ML-KEM, without changing Bitcoin and without requiring the network to upgrade all at once.

## Key Contributions

1. **A post-quantum design for Lightning's off-chain surfaces.** PQLN requires no change to Bitcoin and is built to interoperate with unmodified ("vanilla") nodes, using a hybrid conservative-extension approach in which all existing secp256k1 mechanisms stay in place alongside ML-DSA and ML-KEM.
2. **Gossip as quantum-safe key distribution with downgrade resistance.** Nodes publish ML-DSA and ML-KEM public keys inside their signed `node_announcement`, peers pin them trust-on-first-use, and security decisions depend on pinned or offer-committed keys rather than on forgeable in-band negotiation messages.
3. **Format-level techniques for fitting large post-quantum material into fixed Lightning size limits.** Signature chunking across the 639-byte BOLT 11 tagged fields, and a fixed-size ciphertext list of 20 slots beside the unchanged 1300-byte onion payload, plus a 10-slot list for blinded paths.
4. **Security analysis, implementation and evaluation.** One proposition per threat reducing guarantees to EUF-CMA for ML-DSA-44 and IND-CCA for ML-KEM-768, an implementation by modifying rust-lightning source (publicly released), and measurements of computation, communication, latency and interoperability with real Lightning nodes at every NIST parameter set and with FN-DSA substituted for ML-DSA.

## Main Findings

- **Interoperability holds in practice:** PQLN nodes interoperate with unmodified nodes across the main Lightning operations, from channel opening to invoice, offer, and multi-hop payments, as shown with real Lightning nodes.
- **Computational overhead is small:** Each added cryptographic operation takes at most 0.33 milliseconds, and a post-quantum payment adds less than 60 milliseconds per hop on a typical Internet link.
- **Communication dominates the cost:** A PQLN node downloads about ten times as much gossip data as an unmodified node with ML-DSA, or about four times as much with FN-DSA (the Falcon-based scheme under standardization at NIST, not yet published at the time of writing).
- **Size limits required new encoding tricks:** A 2420-byte ML-DSA-44 signature does not fit into invoice fields of at most 639 bytes each, so it is split across four tagged fields (with three fields for the optional 1312-byte public key); one 1088-byte ML-KEM-768 ciphertext would nearly fill the 1300-byte onion payload, so ciphertexts travel in a 20-slot list beside the onion.
- **Relay policy, not acceptance, blocks post-quantum gossip:** Vanilla rust-lightning accepts, verifies and stores a post-quantum announcement but refuses to relay it when unrecognized trailing data exceeds 1024 bytes; PQLN raises this relay budget to 8192 bytes, so a vanilla node on the path stops propagation without rejecting the message.
- **No in-band negotiation:** The hybrid handshake runs on a dedicated port and the classical port stays byte-identical to vanilla, so a quantum adversary has no negotiation message to rewrite into a downgrade. Pinned or offer-committed keys are the only trust anchors.
- **Offers get their own anchor:** Each BOLT 12 offer commits a per-offer ML-DSA key derived via HMAC of the node's symmetric offer key and the offer's nonce, so invoice verification works even for payees that never appear in gossip; distinct offers carry unlinkable keys.
- **Onion privacy is hybridized without changing the packet:** Per-hop ML-KEM shared secrets are folded into each hop's classical Sphinx secret, and dummy list entries are encoded from uniformly random polynomial coefficients exactly as FIPS 203 builds a ciphertext, so a hop cannot count real entries by frequency.
- **Invoices have a residual unprotected case:** Without a trusted key, an invoice carrying post-quantum fields is "unanchored" and gives no protection, since a signature under a self-asserted key proves nothing; unannounced payees on first contact remain unprotected.
- **Clear scope boundary:** BOLTs 2, 3 and 5 are out of scope because their secp256k1 keys live inside Bitcoin transactions; BOLTs 4, 7, 8, 11 and 12 are protected, and two experimental feature bits are registered under BOLT 9.

## Methodology in Plain English

The authors first map each Lightning specification (BOLT) to the elliptic-curve operation behind it and separate the surfaces that live inside Bitcoin transactions from those that live purely in messages between nodes, since only the latter can be changed by a software update. They then extend each off-chain surface by adding lattice-based cryptography alongside the existing cryptography rather than replacing it, so that even a classical attacker gains nothing and a quantum attacker must additionally break ML-DSA or ML-KEM.

To keep unmodified nodes working, they carry every addition in optional odd-numbered Type-Length-Value records, which Lightning's "it's ok to be odd" rule tells implementations to skip, and they register feature bits only for discovery, never for security decisions. Where the post-quantum material exceeds a hard size limit, they either chunk it across several fields (invoices) or move it into fixed-size parallel lists next to an untouched packet (the onion and blinded paths). Keys are derived from the node's existing seed at two dedicated hardened BIP 32 indices, so an existing wallet backup restores the post-quantum identity too.

For transport, they modify the Noise_XK handshake to fold two ML-KEM encapsulations into Noise's chaining key and transcript hash, and require the initiator to already know the responder's static ML-KEM key from gossip or out of band. They then build the design into rust-lightning, run real Lightning nodes to test interoperability and measure computation, gossip volume and payment latency, and finally write one security proposition per identified threat, each reducing its guarantee to the standard notions assumed for ML-DSA-44 and ML-KEM-768.

## Why This Matters

For research, the paper opens a concrete agenda in a space that previously held only a forum analysis that stopped short of a design: it shows that a second-layer protocol can migrate to post-quantum cryptography ahead of, and independently of, its base layer, and it supplies a reusable pattern for hybridizing Noise, Sphinx and TLV-based message formats under hard size limits.

Real-world applications:

- **Payment channel networks and other second-layer protocols** that inherit elliptic-curve cryptography from an underlying chain and need harvest-now-decrypt-later protection before a consensus fork is feasible.
- **Onion-routing and mixnet systems** facing the same fixed-packet-size constraint, where the paper's sidecar ciphertext list and its frequency-correct dummy encoding are directly transferable.
- **Lightning service operators and wallet providers** who need a deployable upgrade path, since PQLN nodes join today's network without a coordinated flag day and keep working with unmodified peers.
- **Transport-layer security engineering** more broadly, as an example of migrating away from Diffie-Hellman in a handshake without any in-band negotiation surface for an attacker to exploit.

Industry relevance is significant because Lightning is described in the paper as the most widely used payment channel network, with around 16,000 nodes and 33,000 public channels holding roughly 3750 BTC at the time of writing, and because Coinbase enabled Lightning transfers in 2024. Bitcoin dominance over total cryptocurrency market capitalization exceeds half, and around 6.9 million BTC, more than a third of circulating supply, sits in outputs with exposed public keys, which frames the wider urgency. NIST's plan to disallow elliptic-curve signatures and key exchange after 2035 sets a deadline that off-chain infrastructure cannot meet by waiting on a Bitcoin consensus change.

## Future Directions

- **Base-layer and channel-fund coverage.** The secp256k1 keys inside funding, commitment, penalty and HTLC outputs remain out of scope and await Bitcoin's own post-quantum transition; how a Lightning design should interact with post-quantum outputs, or with adaptor-signature and script-level proposals, is left open.
- **Bootstrapping trust-on-first-use safely.** The model assumes keys were pinned before a cryptographically relevant quantum computer exists, since a first contact made afterwards can be intercepted by a live adversary; a practical key-transparency or out-of-band verification mechanism for first contact is not solved here.
- **Gossip bandwidth and relay economics.** The roughly tenfold gossip growth with ML-DSA and fourfold with FN-DSA, combined with the raised 8192-byte relay budget, invites work on compression, aggregation or alternative distribution, and on what happens once FN-DSA's standard is finalized.
- **Standardization and multi-implementation deployment.** Turning the odd TLV records, feature bits, dedicated port and the `build_post_quantum_blinded_paths`, `require_post_quantum_payments` and `require_post_quantum_inbound` flags into agreed BOLT specifications, and repeating the evaluation across other Lightning implementations, are natural next steps.

## Target Audience

Cryptography and protocol engineers working on Bitcoin or Lightning implementations, security researchers studying post-quantum migration of deployed systems, and graduate students in applied cryptography who want a worked example of fitting large lattice-based keys, signatures and ciphertexts into legacy message formats with hard size limits. It also suits technical decision-makers at wallet, exchange or node-operating organizations who need to judge whether an off-chain post-quantum upgrade is deployable ahead of a base-layer consensus change.

## Authors’ abstract

A cryptographically relevant quantum computer will break the elliptic-curve cryptography behind Bitcoin and its Lightning Network, the most widely used payment channel network. Even a post-quantum consensus upgrade of Bitcoin would not cover Lightning's off-chain surfaces, so its gossip, peer transport, invoices, payment onions, and offers need separate protection. An adversary can already record Lightning's encrypted traffic today and decrypt it once such a computer exists.…

[The full abstract on arXiv](https://arxiv.org/abs/2609.13781)

[Read the original paper](https://arxiv.org/abs/2609.13781)

---

Canonical: https://crypto.world/paper/pqln-post-quantum-security-for-the-bitcoin-lightning-network-s-off-2609.13781
