Effective Malware Detection Approach Based on Deep Learning in Cyber-Physical Systems

Cyber-physical Systems based on advanced networks interact with other networks through wireless communication to enhance interoperability, dynamic mobility, and data supportability. The vast data is managed through a cloud platform, vulnerable to cyber-attacks. It will threaten the customers in terms of privacy and security as third-party users should authenticate the network. If it fails, it will create extensive damage and threat to the established network and makes the hacker malfunction the network services efficiently. This paper proposes a DL-based CPS approach to identify and mitigate the malware cyber-physical system attack of Denial of Service (DoS) and Distributed Denial of Service (DDoS) as it ensures adequate decision support. At the same time, the trusted user nodes are connected to the network. It helps to improve the privacy and authentication of the network by improving the data accuracy and Qua lity of Service (QoS) in the network. Here the analysis is determined on the proposed system to improve the network reliability and security compared to some of the existing SVM-based and Apriori-based detection approaches.