Privacy-Preserving E-Voting Scheme with Dynamically Involved Voters
Xin Li, Wenju Xu, Yunfei Cao, Jing Zhang, Wei Zhao, Dianhua Tang · 2024
As a more economical and efficient voting method than traditional voting, electronic voting (e-voting) has been widely concerned and supported. However, e-voting schemes face great challenges in privacy and security protection. In order to solve the problem of privacy protection, we design a multi-key fully homomorphic encryption scheme based on BGV (TOCT 2014), which can well protect data privacy of each voting participant. The scheme converts voters' different keys into the same key for encryption to control the key and ciphertext size, uses re-linearization method with modulus improvement technology to reduce noise, and adopts distributed decryption method to protect data privacy of each voter. Compared with other conventional e-voting schemes, our scheme has relatively small communication cost, sorts candidates with an adapted ciphertext comparison method, and supports multi-candidate voting and dynamically involved voters, so it is more practical and scalable. The security analysis proves that the scheme can also achieve data privacy and security protection. In addition, a simulation experiment of electronic voting scheme based on Helib cryptographic library is designed. The experimental results show that our scheme only takes hundreds of milliseconds to complete the voting in small-scale voting scenarios, which is more flexible, efficient, and practical.