Cyber-Physical Intrusion Detection with Spider Monkey Optimized GRU-LSTM and Blockchain Security

Intrusion detection in Cyber-Physical Systems (CPS) has gained increasing importance with the rise of sophisticated cyber threats. Existing detection methods often suffer from high false. Large positive rates and real-time inefficiency in CPS entails the need for accurate and high scalable intrusion detection systems. In this study, a novel AI-Enhanced Spider Monkey Optimization (SMO) algorithm, integrated with a hybrid GRU-LSTM model, to enhance CPS intrusion detection is presented. Additionally, Blockchain technology is incorporated into the framework to enable decentralized communication between CPS components, preventing data tampering and enhancing overall system security. The proposed system operates in two phases: first, features are extracted using the hierarchical GRU-LSTM model, followed by the selection of optimal features for intrusion detection via the AI-Enhanced SMO. Developed in Python, the model shows promising results with the primary experiments, causing 99.4% accuracy rather than 87% of conventional approaches to have more false alarms.