Classification Method of Industrial Control Protocols Based on Statistical Model Matching and Association Rule Analysis

Due to significant differences among industrial control system (ICS) protocols, including design objectives, communication methods, security performance, and applicable scenarios, this paper proposes a frame header recognition method for classifying industrial control protocols based on statistical models and matching algorithms. This method first employs statistical models to identify candidate frequent items and then combines association analysis algorithms to filter these candidates, ultimately achieving accurate identification of the encapsulation headers of industrial control protocols. Experimental results demonstrate the feasibility and effectiveness of this method.