Mutual Information-based Intrusion Detection Model for Industrial Internet

High dimension, redundancy attributes and high computing cost issues usually exist in the industrial Internet intrusion detection field. For solving these problems, the mutual information-based intrusion detection model for industrial Internet was proposed. Firstly, by using features selection method based on mutual information, the attributes set was reduced and traffic characteristics vector was established. Secondly, the normal and abnormal traffic characteristics maps were obtained via the traffic characteristics map technology based on multi correlation analysis. Finally, with the using of discrete cosine transform and nonnegative matrix factorization, we can produce normal and abnormal hash digest, which were used to produce intrusion detection rules. To verify the effectiveness of this model, we adopt NSL-KDD data as the experimental data. The experimental results show that, by using the features selection approach based on mutual information, the proposed model has good classification accuracy and gets good detection performance.