A Regularized Cross-Layer Ladder Network for Intrusion Detection in Industrial Internet-of-Things

As part of BigData trends, the ubiquitous use of the Internet-of-Things (IoT) in the industrial environment has generated a significant amount of network traffic. In this type of IoT industrial network where there is a large equipment heterogeneity, security is a fundamental issue, thus it is very important to detect likely intrusion behaviors. Furthermore, since the proportion of labeled data records is small in IoT environment, it is challenging to detect various attacks and intrusions accurately. This investigation builds a semi-supervised ladder network model for intrusion detection in IIoT. This model considers the manifold distribution of high-dimensional data and incorporated a manifold regularization constraint in the decoder of the ladder network. Meanwhile, the feature propagation between layers is strengthened by adding more cross-layer connections in this model. On this basis, a random attention-based data fusion approach to generate global features for intrusion detection. The experiments on CIC-IDS2018 show that the proposed approach can recognize the intrusion with less false alarm rate, whilst model training is time-efficient.