In the digital age, the increasing frequency of data breaches and incidents of unauthorized access has highlighted critical weaknesses in conventional authentication systems. This dissertation presents an improved multifactor authentication (MFA) model aimed at enhancing data security and minimizing unauthorized access in both personal and enterprise-level digital systems. The problem addressed stems from the limitations of existing MFA approaches, which often rely on static factors that are susceptible to phishing, social engineering, brute force attack, man-in-the middle attack and device compromise. The study begins by identifying the limitations of existing MFA methods, particularly their reliance on static credentials and lack of adaptability to emerging threat patterns. To address these issues, the research introduces a novel authentication framework that incorporates multiple dynamic factors such as biometric authentication, behavioral analytics, risk-based authentication, and location-based authentication, alongside conventional factors like passwords and one-time codes (OTP).The methodology involved designing the enhanced MFA model, simulating various attack scenarios, and conducting performance evaluations using security metrics such as authentication accuracy, response time, and resistance to breach attempts. Results demonstrated that the proposed model offers improved detection of suspicious activity, faster authentication processing, and significantly reduced risk of unauthorized access. Based on the findings, it is recommended that organizations adopt An Improved Multifactor Authentication Systems to strengthen access control mechanisms. The study concludes with suggestions for further research into integrating artificial intelligence and machine learning to continuously evolve authentication protocols in response to emerging threats.
Bassey et al. (Mon,) studied this question.