Enhancing Security and Privacy in Small Drone Networks Using 6G-IOT Driven Cyber Physical System

The purpose of this abstract is to present a concise summary of a study that sought to enhance the cybersecurity of 6G-IoT drone devices through the use of a layered architectural framework and machine learning methodologies. An improved seven-layered architecture that can classify various types of attacks with a 99% accuracy rate is presented in the paper as a means to enhance drone security and privacy. You may encounter Prob, DoS, R2L, and U2R assaults, among others. On the drone dataset, the RegressionNet model—a combination of Logistic Regression and Multilayer Perceptron—performs best. Impressively, it attains a rate of 99.89% accuracy. Both the STIN security dataset and a combined dataset underwent additional validation testing, with the combined dataset producing an average accuracy of 97.90% and the STIN security dataset 91.64%, respectively, proving that the suggested method is both effective and resilient. By enhancing the cybersecurity of 6G-IoT drones, We can see the importance of the suggested architecture and ML models from these outcomes. This, in turn, guarantees secure drone operations across many fields while reducing privacy issues.