Smartphone’s Multifactor Authentication Using Secret Pattern and User’s Unique Behavioral Habit
Ayahiko Niimi · Journal of Internet Technology and Secured Transaction · 2024
In recent years, the widespread adoption of smartphones has significantly heightened the importance of robust security measures.Multifactor authentication has emerged as a critical method to strengthen authentication processes.This paper examines the application of multifactor authentication on smartphones, focusing on a method that integrates three key factors: 1) possession of a smartphone (possession-based information), 2) knowledge of a password or pattern (knowledge-based information), and 3) biometric authentication, such as fingerprint, face, or voice recognition.Given the possibility that a smartphone may already be in unauthorized possession during an access attempt, combining these factors is essential to enhance security.This study proposes an improved approach to the screen unlocking process through enhanced pattern authentication.Typical pattern authentication requires users to trace a sequence of nine points in a specific order to unlock their device.However, this method is inherently vulnerable if an unauthorized party observes and memorizes the pattern.To address this issue, we introduce a machine learning-based solution that enhances the security of pattern authentication by leveraging the unique behavioral traits of the user.This model analyzes not only the sequence of traced points but also the broader characteristics of the user's interaction with the device as a dataset.We developed a data collection application to support this research and validated the proposed method's effectiveness.