Edinburgh Napier University

  With the continuous evolution of the smart environment, the use of Internet services is becoming more common. The increased usage of these services has impacted the traditional network boundaries. This shift has created a need for more flexible, accurate and balanced approaches to cater to user authentication. Current research assumes that users, devices and processes involved in authentication within a network are trustworthy but lack in preventing attacks and managing security threats effectively. This research addresses this issue and proposes a continuous facial recognition framework based on environmental detection. This research provides a two-way mutual authentication between a user and the server. Unlike older systems, this scheme continuously monitors the user environment and uses continuous biometric authentication to authenticate the user based on the environment changes. The scheme is verified using SVO logic and supported with security and performance analysis.