Versatile unsupervised anomaly detection method for RTE-based networks

作者：

Highlights：

• An Anomaly detection method for RTE-based networks using One-Class SVM is proposed.

• The use of only statistical data features guarantees the generality of the proposal.

• The proposal’s performance is verified using PROFINET and Ethernet/IP traffic data.

• The use of optimal values enhances the performance of the proposal.

摘要

•An Anomaly detection method for RTE-based networks using One-Class SVM is proposed.•The use of only statistical data features guarantees the generality of the proposal.•The proposal’s performance is verified using PROFINET and Ethernet/IP traffic data.•The use of optimal values enhances the performance of the proposal.

论文关键词：OCSVM classifier,Differential evolution,PROFINET,Ethernet/IP,Anomaly detection

论文评审过程：Received 18 May 2021, Revised 10 December 2021, Accepted 1 June 2022, Available online 8 June 2022, Version of Record 15 June 2022.

论文官网地址：https://doi.org/10.1016/j.eswa.2022.117751