Simulation of an Intelligent Multi-Factor Access Control System Using RFID, PIN and Behavioral Authentication

Single-factor identification methods such as RFID cards remain vulnerable to cloning, loss, and unauthorized duplication, which limits their suitability for high-security environments. The objective of this research is to design and simulate an intelligent multi-factor access control system that layers RFID card identification, Personal Identification Number (PIN) verification, and behavioral authentication through keystroke dynamics into a single decision pipeline. The system was modeled as a finite state machine and evaluated entirely in a software simulation environment representative of an Arduino-class embedded controller, using a simulated MFRC522 RFID reader, a 4x4 matrix keypad, and a keystroke-timing profiler. Each authentication attempt was required to pass all three factors in sequence before a simulated relay-actuated lock was released. Simulation trials comprised 40 RFID identification attempts, 30 PIN entry attempts including a brute-force lockout test, and 40 keystroke-dynamics classification attempts using timing vectors generated from simulated legitimate and impostor typing profiles. The findings include a 95% RFID identification success rate for authorized tags with 100% rejection of unauthorized tags, a 100% enforcement rate for the three-attempt PIN lockout policy, a keystroke-dynamics classification accuracy of 92.5% with a false acceptance rate of 5% and a false rejection rate of 10%, and an overall combined-factor system accuracy of 97.5% with an average total authentication latency of 2.4 seconds. The research concludes that layering possession-, knowledge-, and inherence-based factors in a single simulated pipeline substantially reduces the attack surface associated with single-factor RFID access control while remaining feasible for low-cost embedded deployment.