A Review on Fraud Detection Techniques in Iot-Enabled Voting Systems

Abstract

The integration of Internet of Things (IoT) in the electronic voting systems has improved the efficiency, accessibility of the electoral process and transparency of the voting process to a large extent but it has also come with its own security vulnerabilities and risks associated with fraudulent activities. IoT-enabled voting systems rely on interconnected devices, wireless-based communication, and real-time data transfer, which increase the attack surface and expose the system to various risks, including voter impersonation, vote manipulation, replay attacks, denial-of-service attacks, insider fraud, etc. This review paper presents a comprehensive discussion of fraud detection methods used in the IoT-enabled voting systems. It critically analyzes the cryptographic and blockchain-based security solutions, fraud detection models that can utilize machine learning, and hybrid systems that can combine various methods. The research paper provides the advantages and disadvantages of both methods based on their strength in security, scalability, cost of computation, and detection ability. Comparison of the results in tables and graphical representations shows that hybrid fraud detection architectures provide better robustness and detection rates as they are a composite of secure communication, data integrity and smart anomaly detection. Such challenges as scalability, preserving privacy or resource constraints are also highlighted in the review and give the insights on the possible direction of future research about the creation of secure and trustworthy IoT-based voting infrastructures