Detecting and responding to attacks and weather effects in hybrid FSO/RF systems using the Dempster-Schaffer theory with AI algorithms

In this paper, a new intelligent switch for hybridFree-Space Optical (FSO) RF communication is proposed forimproved reliability and security in the presence of dynamicenvironmental changes and cyber-attack interferences. UsingDempster-Shafer Theory (DST) for reliable threat classificationand ANN, KNN, and SVM for machine learning, an extraordinaryreal-time communication link selection is achieved. Abroad training dataset (10,000 simulated samples), coveringeavesdropping and jamming threats, fog and dust effects, wasused to train and validate the network. Our work combinesDST to combine evidence from multiple sources and make anaccurate belief assignment for communication modes. In addition,the system exhibits a high claimed confidence, RF and FSO linkbeliefs around 0.88-0.89 and 0.82-0.83, respectively. The machinelearning models have excellent performance on threat detectionand mode classification. ANN, KNN, and SVM obtained accuraciesof 0.99986, 0.99984, and 0.99930, respectively. All modelsachieved near-perfect AUC values, where most classes reach 1,meaning a better discriminative performance. Importantly, theperformance of ANN was significantly outperformed by KNNand SVM in all metrics, demonstrating its robustness. Thiswork provides an efficient and dynamic approach to keep thecommunication in difficult FSO/RF links secure and reliable,and brightens the path for future communication systems.