Self-Enforcing E-Voting System: Trustworthy Election in Presence of Corrupt Authorities
This project aims to investigate a new type of e-voting system, called self-enforcing e-voting (SEEV), for use in future elections. A SEEV system is End-to-End (E2E) verifiable, but in contrast to all other E2E verifiable voting schemes, it does not involve any Tallying Authorities (TAs), hence the system is "self-enforcing". Our hypothesis is that all other existing E2E voting protocols’ dependence on TAs (who need to be not merely trustworthy, but also have significant cryptographic and computer expertise) is a key obstacle to their practical deployment. In a SEEV system, the need for such TAs is removed by applying novel encryption techniques such that the integrity of the tallying result can be verified without requiring any secret keys. Our achievements include three concrete authority-free solutions that systematically cover all three categories of e-voting: 1) centralized Internet voting; 2) centralized polling station voting; 3) decentralized voting. In the first category, we have proposed a SEEV protocol called Direct Recording Electronic with Integrity (DRE-i). An Internet-based implementation of DRE-i has been built and successfully used for classroom voting and student prize competition with positive feedback. In the second category, we have proposed a different SEEV protocol called Direct Recording Electronic with Integrity and privacy (DRE-ip). This protocol adopts a real-time computation strategy as opposed to the pre-computation strategy in DRE-i, and is particularly suitable for polling station voting. A prototype implementation of DRE-ip based on using a touch-screen tablet as the voting client has been built and successfully trialed in the campus of Newcastle University in July 2017 and will be further trialed in local elections in the UK as well as in India. In the third category, we have presented the first smart contract to support boardroom voting over the Ethereum blockchain with maximum vote secrecy without involving any tallying authorities. Overall, our research confirms the feasibility of SEEV for real-world elections, and when implemented properly, SEEV can prove significantly more secure, efficient and usable than traditional paper ballots.