| [Funded by DLT Science Foundation] Abstract: As the groundwork of Risk Management Framework of Blockchain and DLT Systems (RMF-BDLT), this project focuses on monitoring, analysing and modelling blockchain reward mechanisms to manage both internal risks (inherent in protocol design) and external risks (triggered by strategic behaviour). We will develop extensible modelling of PoS staking, which could benefit the design and upgrade of *PoS protocols. Additionally, we will detect anomalies such as selfish and cartel mining behaviour in real-world PoW systems and model the effect of validator’s deviation on PoS consensus. Objectives: Our long-term objective is to develop a comprehensive risk management framework of blockchain and distributed ledger technologies systems (RMF-BDLT). As the initial milestone, at the ground level of the RMF-BDLT, in this project we will mainly focus on the reward mechanisms that are the crucial incentives for participants in blockchain and distributed-ledger technologies (BDLT)-based systems, motivating them to invest resources and maintain the system’s consensus. However, there are many risks associated with the incentives and consensus mechanisms of BDLT-systems: On the one hand, internal risks stem from protocol design and governance; on the other, external risks arise from factors such as strategic validator behaviour and market fluctuations. By addressing both types of risk, we aim to develop a robust RMF-BDLT that can ensure the security and stability of BDLT systems. This project will contribute to both research and practice in several ways. In terms of research: To evaluate the internal risks in incentive design of various protocols, we will first conduct an overview of as many different platforms as possible using a celebrated blockchain taxonomy method developed by Paolo Tasca and Claudio J. Tessone in 2019 [1]. This will help us to extract the main modules and elements in PoS protocol design by summarising the similarities and differences among many variants of PoS protocols (e.g. DPoS, NPoS, PPoS, LPoS). Ultimately, we aim to develop an extensible framework for modelling the staking and reward mechanism in derivatives of PoS systems. Furthermore, we will compare the functioning of different protocols (such as Ethereum2.0 and Hedera) by empirical analysis the system growth, decentralisation of the transaction network, and fairness of staking and reward distribution, enabling us to understand the real functioning of the platforms and forecast the risks associated with the design of the reward mechanisms. By comparing these different platforms, we can identify the most promising approaches to incentive design. Furthermore, to detect the external risks to reward mechanisms caused by strategic behaviour, this project will firstly identify the attacks such as selfish and cartel mining in various real-world PoW-based systems, such as Bitcoin Gold, Monero, Zcash and Dogecoin (extending previous analyses we developed on Bitcoin [3,4]). Additionally, we will model the strategy and deviation of validator in PoS protocols, including both online and offline behaviour, evaluating their risks to the system’s consensus. In terms of practice, this project has several potential applications: On the one hand, our extensible framework for modelling PoS staking, and reward mechanism could be extended to simulate the specific staking and reward dynamics of different real-world systems. This framework is helpful to understand the effects of participants’ staking selection, system’s reward distribution and fee dynamics on the fairness and decetralisation of the systems. The insights gained from this framework can inform the design and deployment of future PoS-based systems, leading to more secure and sustainable blockchain and DLT systems. By providing a tool for evaluating the effectiveness of different incentive designs, our project can contribute to the practical development and adoption of blockchain technology. On the other hand, our approach to detecting selfish miners and mining cartels provides effective techniques for identifying suspicious participants in PoW mining. With this detection method in place, we can explore the possibility of developing a reputation mechanism that can be applied either within the community or incorporated into the protocols. Such a mechanism would help the PoW-based system to better defend itself against strategic mining attacks, leading to improved security and stability for blockchain-based systems. Furthermore, our agent-based modelling of validators’ strategies in PoS mechanism enables early warning of the risks to the system’s consensus and developing appropriate defence approaches. By leveraging this approach, we can proactively address potential threats to the system’s functioning and ensure the continued smooth operation of blockchain-based networks. |
