In recent years the advancements in the smart infrastructure like smart cities, smart devices and their interconnection with each other as well as with the internet has given rise to deployments of systems called cyber physical systems in both public and private domains. Where this advancement has brought ease and comfort into our life, it has also increased the vulnerability to security attacks. The reason is insecure implementation of such systems and transmission of sensitive data over equally or more insecure networks. As a result, exponential increase in attacks like denial of service, ransomware, and flooding attacks has been observed. The development of security techniques is still in its early stages while at the same time deployment of insecure systems has vastly grown. Due to this reason, it is necessary to focus on the security aspects of cyber physical systems to avoid greater harm, like financial and life loss. Therefore, in an effort to contribute towards the security of cyber physical systems, authors have proposed an anomaly based intrusion detection system for detection of denial of service attack using machine learning. Decision tree classifier has been used to develop the system because it can handle categorical as well numerical data which is compatible with the requirements of a cyber physical system. The proposed approach is evaluated using parameters like accuracy and the paper is then concluded with final remarks on results and future direction of research.
Keywords: cyber physical systems Internet of Things,smart infrastructures,denial of service attack, machine learning
How to cite this article: Ponnusamy V, Bakhshad S, Ping OW, Zahra Ft, Noaman NM, Khan A., Attack Detection in Cyber Physical Systems using Machine Learning. Int J Drug Deliv Technol. 2026;16(2s): 42-55; DOI: 10.25258/ijddt.16.42-55
Source of support: Nil.
Conflict of interest: None