A industrial control protocol recognition method based on CNN-EWA-SVM

In modern industrial control networks, a large number of system devices use various communication protocols for communication, which puts considerable pressure on device access and production data exchange. Accurate recognition of industrial control protocols is one of the effective ways to solve the above problems. Due to its unique advantages in protocol feature extraction, machine learning based protocol recognition methods have received sustained attention. However, traditional machine learning relies on manually designed feature extractors, which have poor performance in some unknown systems. Therefore, this article proposes an industrial control protocol recognition method based on CNN-EWA-SVM. Convolutional Neural Network (CNN) is used to automatically extract effective features of protocols. The Enhanced Whale Algorithm (EWA) optimizes the parameters of the Support Vector Machine (SVM), further improving the accuracy of protocol recognition. The proposed method was evaluated using actual protocol data, and the experiment proved its superiority.