Secure Online Examination with Biometric Authentication and Blockchain-Based Framework

Abstract

E-learning has been carried out all over the world and then online examinations have become an important means to check learning effect during the outbreak of COVID-19. Participant authenticity, data integrity, and access control are the assurance to online examination. The existing online examination schemes cannot provide the protection of biometric features and fine-grained access control. Particularly, they did not discuss how to resolve some disputes among students, teachers, and a platform in a fair and reasonable way. We propose a novel biometric authentication and blockchain-based online examination scheme. The examination data are encrypted to store in a distributed system, which can be obtained only if the user satisfies decryption policy. And the pieces of evidence are recorded in a blockchain network which is jointly established by some credible institutions. Unlike other examination authentication systems, face templates in our scheme are protected using a fuzzy vault and a cryptographic method. Furthermore, educational administrative department can determine who the real initiator of malicious behavior is when a dispute arises using a dispute determination protocol. Analysis shows that no central authority is required in our scheme; the collusion of multiple users cannot obtain more data; even if the authorities compromise, biometric features of each user will not be leaked. Therefore, in terms of privacy-preserving biometric templates, fine-grained access, and dispute resolution, it is superior to the existing schemes.

A Comprehensive Survey of Context-Aware Continuous Implicit Authentication in Online Learning Environments

“…The transition from face-to-face learning and teaching to online learning (E-learning) or blended learning has been the subject of considerable interest over the last couple of decades [1] and it has been accelerated recently by the COVID-19 pandemic [2,3]. In the current climate, online learning and teaching have become common practice.…”

“…Most online learning management systems allow learners to log into their own course through authentication. Different authentication mechanisms can be applied in online learning environments including (1) knowledge based (what a user knows), (2) possession based (what a user has) and (3) biometrics based (what a user is) [3]. Knowledge based authentication using usernames and passwords (also known as password-based) is the most widely applied authentication method, but passwords can potentially be vulnerable and stolen, leading to a failure to authenticate the genuine learner [8][9][10].…”

“…Compared to password-based authentication, biometric techniques such as face or keystroke recognition can reduce the issues of credential sharing and attacks (i.e. shouldersurfing attacks) while achieving high authentication accuracy [3,4]. Although biometric systems represent a robust method to authenticate users, challenges have been reported.…”

Ensuring Academic Integrity and Trust in Online Learning Environments: A Longitudinal Study of an AI-Centered Proctoring System in Tertiary Educational Institutions

“…[41]; distributed ledger technologies, having certain features that can address several challenges associated with preservation of the privacy of biometric data. The single point of failure can be addressed by its distributed nature, elimination of third parties and potential privacy leakage, and monitoring and access to trusted and unalterable history logs [40,[42][43][44][45][46].…”

A Reliable Block-Chain Based Biometric Authentication Solution for Aadhar

“…Major security flaws in biometric authentication systems include the possibility of biometric information leakage, the unreliability of authentication modules, and the lack of openness in the handling of biometric information. Because each person's biometrics are distinct and must be maintained securely, and replication of a biometric template is difficult, its application systems are more secure than conventional techniques like passwords (1) . Since biometrics are immutable, consecutive security hazards may result from a biometric data breach (2) .…”