Visible to the public A Theoretical Model for Fork Analysis in the Bitcoin Network

TitleA Theoretical Model for Fork Analysis in the Bitcoin Network
Publication TypeConference Paper
Year of Publication2019
AuthorsShahsavari, Yahya, Zhang, Kaiwen, Talhi, Chamseddine
Conference Name2019 IEEE International Conference on Blockchain (Blockchain)
ISBN Number978-1-7281-4693-5
Keywordsbitcoin, Bitcoin consensus, Bitcoin network, block propagation delay, block selection rules, Blockchain fork, blockchain networks, composability, compositionality, computational complexity, cryptocurrencies, cryptocurrency, cryptography, Erdös-Rényi random graph construction, fork analysis, fork branches, fork occurrence probability, graph theory, historical Bitcoin data, Nakamoto consensus, network bandwidth, network protocols, network simulator OMNET, Peer-to-peer computing, peer-to-peer network, probability, Protocols, pubcrawl, random processes, theoretical cryptography, Theoretical modeling

Blockchain networks which employ Proof-of-Work in their consensus mechanism may face inconsistencies in the form of forks. These forks are usually resolved through the application of block selection rules (such as the Nakamoto consensus). In this paper, we investigate the cause and length of forks for the Bitcoin network. We develop theoretical formulas which model the Bitcoin consensus and network protocols, based on an Erdos-Renyi random graph construction of the overlay network of peers. Our theoretical model addresses the effect of key parameters on the fork occurrence probability, such as block propagation delay, network bandwidth, and block size. We also leverage this model to estimate the weight of fork branches. Our model is implemented using the network simulator OMNET++ and validated by historical Bitcoin data. We show that under current conditions, Bitcoin will not benefit from increasing the number of connections per node.

Citation Keyshahsavari_theoretical_2019