Intrusion Detection for Time-Sensitive Industrial Control Systems

Abstract

Industrial control system (ICS) is specialized control systems used in various industrial processes, such as monitoring, managing, and controlling physical equipment and processes, which is critical for boosting automation and optimizing industry processes. ICS plays a very important function and role in smart manufacturing and Industry 4.0 industries because it can receive remote sensor information through the network, issue instructions to the control station and perform management responsibilities to improve production efficiency and effectively manage all production devices within a factory. However, as more and more devices are connected to the Internet, the potential for cyberattacks increases. On the other hand, traditional Ethernet does not take into account the need to ensure that the transmission of specific data flow is completed within the time constraint in industrial control systems. Industrial control systems will have irreversible consequences if the data cannot be delivered within the deadline. In order for vital traffic from industrial control systems to reach the control end station on time, time-sensitive networks (TSNs) can guarantee that important traffic satisfies predefined low-latency transmission characteristics. As a result, TSN has drawn a lot of interest and been quickly used in industrial control systems. Intrusion Detection System (Intrusion Detection System) is able to identify network attacks with the functioning of important ICS facilities. The existing intrusion detection technology is no longer able to identify attacks in the environment where the networks of ICS and TSN are integrated. Thus, in order to design and implement an IDS appropriate for ICS in a TSN network environment, we proposed an intrusion detection system based on the concept of packet size and period, named Packet size & Period Intrusion Detection System (PsPIDS). Through our analysis, we find that the packet size and packet period scheme is capable of detecting malicious attacks in encrypted environments. The experiment results demonstrate the ability and capability of the proposed IDS, which unquestionably delivers superior security for industrial applications.