Strategic Analysis to defend against Griefing Attack in Lightning Network

Payments routed in Lightning Network are susceptible to a \emph{griefing attack}. In this attack, the channels get blocked, and the affected parties cannot process any payment request. Our work is the first to analyze griefing attacks in Hashed Timelock Contract or \emph{HTLC}, from a game-theoretic point of view. Using the same model, we analyze another payment protocol Hashed Timelock Contract with Griefing-Penalty or \emph{HTLC-GP}, which was proposed to counter griefing attacks. We find that \emph{HTLC-GP} is \emph{weakly effective} in disincentivizing the attacker. To further increase the cost of attack, we introduce the concept of \emph{guaranteed minimum compensation} and integrate it into \emph{HTLC-GP}. This modified payment protocol is termed $\textrm{HTLC-GP}^{\zeta}$ and unlike \emph{HTLC-GP}, the protocol considers the participants to act rationally. By experimenting on several instances of Lightning Network, we show that the capacity locked drops to $40\%$ in the case of \emph{HTLC-GP} when the rate of griefing-penalty is set to $4.5\times 10^{-5}$, and $28\%$ in the case of $\textrm{HTLC-GP}^{\zeta}$ when guaranteed minimum compensation is $2.5\%$ of the transaction amount. These results justify our claim that $\textrm{HTLC-GP}^{\zeta}$ is better than \emph{HTLC-GP} to counter griefing attacks.