Blockchain Using Proof-of-Interaction
Jean-Philippe Abegg, Quentin Bramas, Thomas Noël · Lecture notes in computer science · 2021
Proof-of-Work is originally a client-side puzzle proposed to prevent spam or denial of service attacks. In 2008, Satoshi Nakamoto used it as an election mechanism (or equivalently, to replace a centralized time server) in the first Blockchain: Bitcoin. In the same year, another spam prevention algorithm was proposed, based on a guided-tour puzzle, but received only little attention.The main motivation of our work is to see if a Blockchain protocol can use the guided-tour puzzle like Bitcoin uses Proof-of-work.In this paper we extend the guided tour puzzle to a new Puzzle called Proof-of-Interaction and we show how it can replace, in the Bitcoin protocol, the Proof-of-Work algorithm. We show that it uses a negligible amount of computational power compared to Bitcoin, and scales very well in term of number of messages. We analyze the security of our protocol and show that it is not subject to selfish mining. However, our protocol currently works only when the nodes in the network are known, but we discuss how this assumption could be weakened in future work.