MODELING BLOCKCHAIN ATTACKS AND MECHANISMS OF THEIR COUNTERACTION USING MULTI-AGENT SYSTEMS

Abstract

The growing interest in blockchain and cryptocurrencies is driving the research and use of attacks that can target different parts of the system. For example, some attacks target the network or the system itself through sharding, DDOS attacks, or wormhole attacks. Some other attacks, such as the 51% attack, aim to acquire a majority share of the production capacity in the system. With these powers, an attacker or group of attackers can rewrite the blockchain at will, perform double-spend attacks, or censor any entity they choose.

Other attack types combine multiple attack vectors into one, building a competitive chain using unknown participants that were previously isolated from the network via an obfuscation attack. Other attack methods also exist through exploitation of smart contracts/decentralized applications bugs. Most decentralized applications are susceptible to favouritism attacks, where an attacker unfairly exploits information related to events that have not yet been recorded on the blockchain.

Provided list of attack vectors, despite its inexhaustibility, is aimed to emphasize the fact that blockchain systems have vulnerabilities. The presence of the latter indicates the need to promote detailed information about blockchain security mechanisms through their simulation.

This article is designed to explore the possibilities of modeling the most common types of attacks on the blockchain by using an organization-oriented modeling method, as well as to show in practice the method of modeling regulatory mechanisms through the reproduction of special meta-agents – regulatory agents – responsible for the coordinated operation of the entire system, and in particular contributing to resistance attacks on the network by checking the proposed transactions for compliance with the rules laid down in them, which results in endorsement or rejection of the former.

Further research may involve modeling the motivation of system agents in the event of rejection of transactions by regulatory agents.