Lightning Network Quantum Resilience: The Payment Channel Lifecycle as a Measurable Risk Surface of a Second-Layer Payment Protocol Built on Top of the Bitcoin Blockchain

02.11.2026

We explore how empirical data on the lifecycle of Lightning Network payment channels can be transformed into a measurable basis for assessing and planning Bitcoin’s post-quantum migration. The paper by Grötschla et al. (arXiv:2409.15930v1) collects data on channel opening, lifetime, and closure, cooperative and unilateral closures, unresolved HTLCs at closure, channel imbalance, and output reuse. We show that these metrics are not only a description of the network but also  a map of the locations where cryptographic keys and channel states become observable on the blockchain . Based on this, we formulate five testable hypotheses for prioritizing post-quantum migration, propose a protocol for the next experiment, and explicitly separate measured facts from future risk assessments.

1. Statement of the problem and intrigue

⚙️ Detailed research option

The Lightning Network is layer two on top of Bitcoin: payments occur off-chain, while the transactions that open and close channels remain primarily on the blockchain. This raises an intriguing question:  if a future quantum computer can forge signatures, where exactly is the measurable “attack surface” in the Lightning Network, and can it be estimated from the data already published on-chain?

The PDF shows that public and private channels differ significantly: private channels are often short-lived, more imbalanced, and less frequently announced in the gossip network. This is important for quantum computing: public channels reveal nodes and parameters via gossip, while closures reveal witness data and public keys. Private channels are more secretive, but their closures still leave on-chain traces.

The central question of the paper is:  is it possible to construct a testable model of post-quantum migration priorities using Lightning Network channel lifecycle data without misrepresenting the risk assessment as a proven attack?

Hypothesis:  The metrics of closures, unresolved HTLCs, and output reuse allow us to rank channels and periods by the degree of observability of cryptographic materials and, therefore, by the priority of migration measures.

Refutation criterion:  If, after annotating key events, it turns out that these indicators do not correlate with the moment of disclosure of public keys/witness data and do not change migration priorities, the hypothesis should be rejected or significantly limited.

2. Mechanism: What exactly is involved in the quantum threat?

⚙️ Detailed research option

Bitcoin uses elliptic curve digital signatures (ECDSA and Schnorr). For a hypothetical cryptographically significant quantum computer (CRQC), Shor’s algorithm is considered a threat to the discrete logarithm and factorization problems. For hash functions, the picture is fundamentally different: Grover’s algorithm provides quadratic, not exponential, speedup, so hashing doesn’t disappear “with a single click.”

In Lightning, a channel is represented by a P2WSH output with a 2-of-2 multisig. Until the output is spent, the script is hidden by the hash. When a channel is closed—especially one-way—commitment transactions, witness data, local and remote outputs, HTLCs, and anchor outputs appear on the blockchain. The PDF directly classifies these outputs and analyzes whether the commitment has been revoked. In the quantum framework, this means that  channel closure is an event of cryptographic disclosure , not simply an economic action.

Moreover, the public keys of public channel nodes can already be known from gossip messages  channel_announcement and  node_announcement. Private channels are not announced, but their funding and closing transactions are still visible on-chain. Therefore, the correct question is not “is Lightning hacked?” but “what data and at what point becomes available to an adversary, according to the threat model.”

3. What is proven in the PDF and what is our interpretation

⚙️ Detailed research option

StatementStatusWhat does this mean for the quantum topic?
Public channels live on average 143 days, private ones – 125 days.Measured in PDFMigration windows are short: old channels close and reopen, creating repeated opening events.
More than 50% of public closings are cooperative; of these, 60% have two exits, 40% have one.Measured in PDFCooperative closure is a potentially convenient point for the implementation of post-quantum procedures, but requires the consent of both parties.
Unilateral closures account for slightly less than half of public closures; rare revocations: 103 public, 78 private.Measured in PDFA one-way path reveals commitment and witness data; rare reviews do not eliminate the need to analyze key material.
More than 60% of cooperative and more than 70% of one-way public channels have a side with ≥95% of the funds at closing.Measured in PDFThe imbalance concentrates economic risk: the “empty” side may be less interested in the security of closure.
20,804 unresolved HTLCs in public and 364 in private closures; a total of 21,168 observations.Measured in PDFThis is the first available set of HTLC closing fee sizes; it allows one to estimate the amount of unfinished state, not just the number of channels.
35% of closed public channels finance at least one other public channel; among cooperative closed public channels, the figure is 56%.Measured in PDFReuse of outputs is an indicator of “peeling chains” and at the same time a natural point for migration audit.
A quantum computer can forge Bitcoin signatures given the right resources.External hypothesis, not measurement PDFRequires a resource model, an error model, and a threat model; this is not a conclusion of this paper.

4. Deep PDF Data Analysis

⚙️ Detailed research option

4.1. Opening Channels

⚙️ Detailed research option

The PDF data covers January 1, 2019, through September 23, 2023, with shortened windows for some analysis. In 2020, there were approximately 5,000 openings per month; in 2021, up to 15,000; private channels accounted for approximately 22% of openings. Channel sizes range from 10 to  10 satoshis  , with private channels being smaller on average. For quantum migration, this means a heterogeneous ecosystem: different sizes, different operators, different horizons.

4.2. Lifetime and gossip activity

⚙️ Detailed research option

The number of active public nodes grew to approximately 12,500 in early 2022 and then fell to just over 7,500 by mid-2022. The number of public channels exceeded 45,000 in early 2022; private channels have always made up less than 20% of public channels. The average number of daily channel updates was 0.69, the median was 0.05; only 8.9% of public channels updated at least once a day on average. This indicates a highly skewed distribution: a minority of channels actively participate in routing.

An analysis of proportional fees revealed a negative correlation between the channel’s parties: the parties attempt to direct the flow in opposite directions and balance liquidity. This is important for post-quantum migration: operators are already using economic signals, meaning migration incentives can be built into the same fee and update mechanisms.

4.3. Closures and imbalances

⚙️ Detailed research option

The PDF distinguishes between cooperative closures ( coopx1,  coopx2) and unilateral closures ( local,  local+remote,  remote,  revoked). For public channels, cooperative closures account for over 50%; unilateral closures account for just under half. The share is  remote around 10%—a surprisingly high figure, considering that the party not receiving funds still initiates the closure. Feedback is extremely rare: 103 public and 78 private for the entire period.

The imbalance is measured by a standardized indicator:

I = 2 max(out₁, out₂) / (out₁ + out₂) − 1

where  I=0 is the total balance,  I=1 is all the funds held by one side; for one exit, the other is considered zero. In PDF: the average imbalance of cooperative public channels is 0.79, unilateral ones – 0.87; private ones – 0.88 and 0.96, respectively. More than 60% of cooperative and more than 70% of unilateral public channels had a side with ≥95% of the funds; for private ones – more than 75% and more than 90%. This means that channels usually close not “in equilibrium,” but when one direction is exhausted.

4.4 HTLC as the first payment dataset

⚙️ Detailed research option

Unresolved HTLCs at the time of closure provide a rare window into payment sizes. 20,804 HTLCs were found in public closures, while 364 were found in private ones, for a total of 21,168. 96% of public and 99% of private channels have no unresolved HTLCs; the remaining private ones have exactly one. The average HTLC size is 230,000 satoshis in public channels and 360,000 satoshis in private channels. Peaks are observed at round amounts of 1,000, 2,000, 3,000, and 10,000 satoshis, with slight excess due to fees.

HTLCs are important for quantum computing for two reasons. First, they represent incomplete states that, if unilaterally closed, could end up on the chain. Second, their sizes allow us to estimate the economic value of states, not just their number. However, the PDF does not prove that every HTLC reveals a quantum-vulnerable signature; this must be verified separately using witness data.

4.5. Reusing Outputs

⚙️ Detailed research option

For 35% of closed public channels, at least one output was used to fund another public channel; among cooperatively closed public channels, the figure is 56%. For private channels, the figures are 14% and 33%, respectively. This points to “peeling chains” and the fact that closure is often not the end, but a reopening. From a migration perspective, these are natural checkpoints: if an operator reopens a channel, they can choose a post-quantum format, provided the protocol is compatible.

5. Quantum Threat Model: What to Test

⚙️ Detailed research option

We define a simplified threat model. The adversary has CRQC and can forge signatures for public keys they know. They don’t automatically obtain private keys, don’t break hash functions exponentially, and don’t bypass authentication protocols without introducing additional vulnerabilities. The model distinguishes between:

  1. The public key is already known  – for example, from gossip for a public node.
  2. The public key is revealed when the output  witness is spent in the closing transaction.
  3. The channel state is not complete  – unresolved HTLCs, commitment exits, timelocks.
  4. Economic incentive  – the imbalance and size of funds influence who will respond and when.

Post-quantum NIST FIPS standards 203, 204, and 205 (August 2024) define key establishment and digital signature algorithms, but their implementation in Bitcoin and Lightning requires separate protocol design: signature sizes, script compatibility, transaction fees, wallet and router support. Therefore, channel lifecycle measurement is not a substitute for cryptanalysis, but a way to prioritize where migration will yield the greatest impact.

6. Testable hypotheses

⚙️ Detailed research option

HypothesisExaminationWhat will refute
H1. The proportion of channels with key material disclosure at closing is higher among unilateral ones than among cooperative ones.Annotate closing transactions by witness and exit type; compare shares.The difference is statistically insignificant or the sign is opposite.
H2. Unresolved HTLCs are concentrated in channels with high imbalance.Joint HTLC analysis and  Iregression/contingency tables.The connection disappears after controlling the channel size and closure type.
H3. Exit reuse is more common in cooperative public closures than in unilateral ones.Check 35%/56% on full window and break down by closing type.The difference is explained by the size of the channel or time.
H4. Private channels have higher imbalance and shorter lifetimes, which reduces the window for migration procedures.Compare distributions, not just means; account for censoring.The differences are not robust to window selection and privacy heuristics.
H5. Life cycle indicators can be used as a priority for post-quantum audit without deanonymization.Aggregated risk rating; check if it restores identities.The rating allows you to deanonymize specific channels/nodes.

7. Connection with the research questions framework

⚙️ Detailed research option

The following questions were selected from the  quantum_research_questions_ru.html constructor  that are directly related to the topic: 1–6 (the problem and refutation), 7–12 (the mechanism), 13–18 (the history and the debate), 19–24 (the hardware and scale as a framework for future resource assessment), 26 (the input data), 31–35 (errors and reliability), 37–41 (fair comparison), 43–46 (proof and reproducibility), 49–52 and 55–56 (cryptography and the threat of Bitcoin), 63–67 (the value and risks), 69–72 (open questions and future experiments). Questions 19–24 do not mean that we measured a quantum computer; they provide a checklist for future resource assessment.

8. Multiple Line Matrix PDF

⚙️ Detailed research option

PDF LineWhat to extractHow to connect with the quantum themeWhat to check
Opening channelsNumber, size, publicityWhere new migration objects are createdAre opening windows compatible with post-quantum scripts?
Life time143/125 days, gossip updatesExposure window length and reopening frequencyWill the migration be completed before the closure?
ClosingsCooperative/unilateral, revocationsThe moment of revealing the witness and keysWhat types of closures are critical?
HTLC21,168 observations, sizesUnfinished states and economic valueHow many HTLCs actually expose vulnerable materials?
Reuse35%/56% public, 14%/33% privateNatural points of migration auditDoesn’t auditing create deanonymization?

9. Honest comparison and limitations

⚙️ Detailed research option

  • The PDF does not contain quantum measurements; all quantum statements here are hypothesis statements and evaluation frameworks.
  • PDF data is aggregated; private channels are determined by heuristics, not absolute truth.
  • Gossip data contains gaps; some analysis is limited to January 1, 2020 – July 1, 2021 or July 1, 2022.
  • We did not perform any new on-chain measurements or check the witness data for quantum-vulnerable signatures.
  • CRQC resource estimates are highly dependent on the architecture, error correction, and adversary model; precise numbers without such context would be unscientific.
  • NIST post-quantum standards do not equal ready-made support in Bitcoin/Lightning; engineering and consensus work is needed.

10. Practical value and economics of migration

⚙️ Detailed research option

The practical value of this approach lies in prioritization. Instead of the abstract “everything needs to be migrated,” channels can be ranked by: (1) the likelihood of key material being compromised, (2) economic size, (3) imbalance, (4) reopening frequency, (5) the proportion of unresolved HTLCs. This doesn’t replace auditing, but it helps plan the compatibility of wallets, routers, and closing procedures. The total cost of ownership includes not only new signatures but also increased transaction size, fees, latency, and the risk of incompatibility.

11. The next experiment

⚙️ Detailed research option

  1. Resample PDF: funding/closing transactions of public channels and heuristics of private channels.
  2. For each closure, annotate: closure type, witness presence, output type, HTLC presence, timelock, output reuse.
  3. Enter the “key material observability index” and test H1–H5 on aggregated data.
  4. Add a model of post-quantum signatures: size, cost, compatibility; evaluate how priority changes.
  5. Conduct a de-anonymization check: any rating should remain aggregated and not reveal specific private channels.

12. Checklist before publication

⚙️ Detailed research option

  • Separated PDF measurements from quantum hypotheses.
  • The existence of a ready-made quantum attack is not claimed.
  • Data limits, privacy heuristics, and gossip passes are specified.
  • Comparing post-quantum circuits is not a matter of “faster/slower” without taking into account size and compatibility.
  • Ethical restrictions: aggregated data only, no de-anonymization.

13. Conclusion

⚙️ Detailed research option

The Lightning Network represents a rare case where the channel lifecycle has already been measured in sufficient detail to serve as a risk surface map for future post-quantum migration. The PDF demonstrates that channels are short-lived, often imbalanced, close cooperatively and unilaterally, and sometimes reopen, while HTLCs provide the first set of outstanding payment sizes. Our paper transforms these facts into a testable framework: which events reveal key information, where the risk is economically concentrated, and which migration measures will have the greatest impact. This is not a promise of a quantum breakthrough, but an exploratory route from measurement to security.

Sources

⚙️ Detailed research option

  1. Grötschla F., Heimbach L., Richner S., Wattenhofer R. On the Lifecycle of a Lightning Network Payment Channel. arXiv:2409.15930v1. PDF data and conclusions used.
  2. National Quantum Initiative. URL:  https://www.quantum.gov/national-quantum-initiative/ .
  3. NIST. FIPS 203, 204, 205 approved, August 2024. URL:  https://csrc.nist.gov/news/2024/postquantum-cryptography-fips-approved .
  4. NIST NCCoE. Migration to Post-Quantum Cryptography. URL:  https://www.nccoe.nist.gov/applied-cryptography/migration-to-pqc .
  5. Lightning BOLT #3: Bitcoin Transaction and Script Formats. URL:  https://github.com/lightning/bolts/blob/master/03-transactions.md .
  6. Poon J., Dryja T. The Bitcoin Lightning Network: Scalable Off-Chain Instant Payments, 2015.