Abstract

We leverage blockchain technology for malicious node detection in Internet of Drone Things (IoDTs). IoDTs additionally provide a secure mechanism of service delivery based on cascade encryption and feature assessment. By making localization and service delivery more challenging, malicious nodes deter new users. The identification and removal of such nodes is necessary for establishing trust between entities. Using federated learning, the proposed system identifies malware. Using SVM and RF classifiers, federated learning is capable of identifying malicious nodes. For the purpose of identifying malicious nodes, we examine their transparency and latency. Nodes that provide services to others provide one another incentives. During service delivery in IoDTs, the challenges that must be addressed are provider and consumer mistrust. These concerns are addressed via feature assessment and cascade encryption. The digital signature used in cascade encryption renders the service provider irrefutable. Due to feature evaluation, consumers get their requested services without fail. Functional testing ensures that a service fulfills industry requirements. Our simulation results demonstrate that our non-spurning model works as designed. Accuracy, recall, and overall F1 score are compared between SVM and RF classifiers. The accuracy of SVM is 79%, its F1 score is 0.88, and its recall is 0.78. The RF classifier has an accuracy rate of 95%, a precision rate of 0.92, an F1 score of 0.96, and a recall rate of 1. Utilizing RF significantly improves the precision of identifying malicious nodes.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call