Biometric Properties and Network Technologies as Determinants for a Secure Electronic Voting Architecture

Abstract

lectronic voting should be uniform, confidential, secure and verifiable as eligible voters are is authenticated by his/her unique characteristics and legible voters not allowed to cast more than one vote. Voting auditors can check whether all ccast ballots are used in computation of the final tally. Most of the e-voting architectures used currently to identify and verify a voter uses single biometric source mainly the fingerprint. Single biometrics source has many problems which include noisy data, intra class disparity, inter class resemblances, universality, spoofing and insecurity. Multi-biometrics sources use multiple source of information for individual authentication. One of the gaps that exist in E-voting implementation in developing countries is the lack of spread of Internet connectivity and electricity. This gap provides a foundation for the research in which an e-voting architecture will be developed based on a polling station setup involving a local area network that uses 3G and 4G technology and General Packet Radio Service (GPRS) for data transfer. This study will evaluate existing electronic voting architectures, investigate biometric technologies in real-world applications, determine existing network technologies necessary for designing e-voting architecture and finally a secure electronic voting architecture will be developed. This research will be carried out using selected counties in Kenya for election processes. The research uses qualitative and quantitative methods. Random and purposive sampling techniques will be used. Data will be collected using observation, interviews, questionnaires, and documents scrutiny. Validity of content and research instruments will be enriched by contributions of specialists, lecturers and supervisors from the department of Information technology. Descriptive statistics will be used to summarize and present data and results. Inferential statistics will be used in making inferences. Reliability of research instruments will be ensured using the split-half method. The findings of this study are significant to the counties under study and to the government in enhancing use of e-voting to improve transparency. An e-voting prototype will be developed to validate the architecture. This architecture will enable developing countries use an e-voting architecture with minimal Internet connectivity using long life batteries where electricity is not available. It will also eliminate irregularities in voter identification, vote casting, and vote counting, vote tallies and auditing