Unjamming Lightning
summary
Proposes a hybrid defense combining local per-link reputation with small unconditional upfront fees (target ~1% of success fee) plus resource bucketing and an HTLC endorsement bit (1/0 TLV) that propagates only when peers trust each other. Reputation is per-HTLC and per-channel: fees forwarded minus an opportunity-cost penalty for slow resolutions (90s blocks), so attackers must pay as much or more to gain trust than the damage they can cause. Paper is a draft specification under development, not yet deployed; notes $240M on LN vs a ~$3M capital jamming estimate.
AI-written summary · about
chapters (11)
- 0:00Jamming attacks on Lightning
- 1:04Channel jamming attack
- 8:33Channel jamming consumes resources
- 13:48Fast versus slow jamming
- 15:55History of jamming proposals
- 21:31Hybrid mitigation solution
- 26:27Resource bucketing and signals
- 31:36Why solution matters
- 34:02Privacy in reputation
- 38:10Reputation algorithm design
- 44:58Summary
related talks
34mState of Lightning Privacy
50mSecurity Aspects of Lightning Network Routing
28mThe Path Forward For Lightning
22mThis One Math Trick Powers the Future of Lightning - PTLCs 101
01h 03mLightning is Broken AF (But We Can Fix It)
48mLightning Network Economics: Cost-Minimal Channels and their Implications for Network Structure
34mUpcoming L2 Systems
01h 00mScaling Bitcoin Off-Chain