University of Illinois at Chicago
Browse

File(s) under embargo

1

year(s)

2

month(s)

2

day(s)

until file(s) become available

Mass Exit Attacks on the Lightning Network

thesis
posted on 2023-05-01, 00:00 authored by Cosimo Sguanci
The Lightning Network (LN) is currently the most popular scaling solution for the Bitcoin blockchain, consisting of payment channels between nodes holding different amounts of BTC. The security of the LN relies on the ability of honest users to close a channel in case of fraudulent behavior by settling their balances, which requires having a transaction confirmed on the Bitcoin blockchain within a pre-agreed time period. We study the susceptibility of the LN to attacks that seek to exploit periods of high congestion on the Bitcoin blockchain and the consequent delay in confirming LN users' transactions. In this scenario, a small coalition of adversarial nodes forces a large set of honest protocol participants to interact with the blockchain: we formulate the problem of finding a worst-case coalition of k adversarial nodes as a graph max-cut problem. We focus on two types of attacks. In the first one, a set of k nodes becomes unresponsive with the goal of locking the funds of many channels for a period of time longer than required by the LN protocol. In the second type of attack, a coalition of k nodes attempts to steal funds by submitting many closing transactions that settle channels using expired channel states. We perform our analysis via simulations that make use of historical data about unconfirmed transactions in the Bitcoin mempool, in periods of high congestion of the blockchain. We show that, under historically-plausible congestion conditions, both attacks can be carried out by a small coalition. These findings highlight the importance of watchtower services, which monitor channels to detect channel breaches and react accordingly: these services must adopt sophisticated strategies to avoid the loss of users' funds.

History

Advisor

Sidiropoulos, Anastasios

Chair

Sidiropoulos, Anastasios

Department

Computer Science

Degree Grantor

University of Illinois at Chicago

Degree Level

  • Masters

Degree name

MS, Master of Science

Committee Member

Medya, Sourav Di Nitto, Elisabetta

Submitted date

May 2023

Thesis type

application/pdf

Language

  • en

Usage metrics

    Categories

    No categories selected

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC