Privacy Preserving Decision-Making over Blockchain

Zhang, Jiajie and Zhang, Bingsheng and Gouglidis, Antonios (2024) Privacy Preserving Decision-Making over Blockchain. PhD thesis, Lancaster University.

Text (2024Jiajie_Zhang_PhD)
Jiajie_Zhang_Thesis_Final_260824.pdf - Published Version
Download (4MB)

Abstract

Decentralised decision-making systems are now a crucial building block of various applications. These systems allow participants to collectively decide the direction and outcomes of their projects, reaching decision agreements in a democratic and inclusive manner. In this thesis, we present our work on a comprehensive and inclusive decentralised collaborative decision-making system while ensuring privacy. The system core foundation is a 2-stage voting scheme based on choice architecture principles. Crucially, this decisionmaking system is built to be compatible with existing blockchains and their infrastructure so it is adaptable and easy to deploy. Furthermore, this system implements liquid democracy and delegative voting, allowing the stakeholders to vote directly on proposals or delegate their voting powers to experts. The combination of wisdom of the crowd and expert knowledge enhances collaborative intelligence, resulting in more informed, and therefore improved decision-making. To ensure the privacy of voters, our system ensures that even when a minority of the voting committee members are dishonest, it is computationally impossible for any participant to reveal the voting preferences or delegations of voters with a significant probability. These privacy assurances are crucial to preserve the integrity of the decision-making process and protect participants. Concurrent multiple voting events is an important feature, therefore in this thesis we introduce a distributed batch key generation protocol. This protocol allows participants to generate multiple keys simultaneously, thereby minimizing communication costs with an amortised complexity of O(n) per key, where n represents the number of participants. Additionally, the system is built to support an evolving committee feature, which allows voting committee members to be changed during the voting process. This ensures that the decision-making process remains adaptable and aligned with the evolving interests and expertise within the blockchain community through flexibility. We thoroughly analyse the security of our system and demonstrate its resilience under the universally composable (UC) framework. By conducting a deep investigation of previous systems, we identify gaps such as non-private ballots and insecure and/or inefficient voting methods which we address in our proposed system successfully. We validate the efficiency of our proposed system and implement a demonstration written using the programming language Scala. The system is then benchmarked, and the results indicate that our system can effectively handle large numbers of participants, while maintaining high efficiency throughout the decision-making process. We believe that this community-inclusive decentralized collaborative decision-making system with privacy assurance contributes to the advancement of blockchain governance. This system also ensures that the principles of decentralization and democratic decisionmaking are upheld. Our proposed system provides incentives for the participants while also ensuring their active engagement in the decision-making process by using a novel reputation management scheme. Our research provides a practical and efficient solution for blockchain applications requiring transparent and secure decision-making processes.