A Security Framework for IoT Using Machine Learning
Abstract
The security of the Internet of Things is attracting increasing attention from both academia and industry. In fact, IoT devices are vulnerable to various security attacks ranging from denial of service (DoS) to network intrusions and data leaks. This paper introduces a novel machine learning (ML) -based security framework that automatically handles the growing security issues associated with the IoT domain. This framework uses both software-defined networking (SDN) and network functions virtualization (NFV) enablers to counter a variety of threats. This AI framework combines an AI-based monitoring agent and response agent using machine learning models broken down into network pattern analysis, along with anomaly-based intrusion detection in IoT systems. The framework uses supervised learning, the distributed data mining system, and the neural network to achieve its goals. In particular, the propagation of attacks using the data mining approach is very successful in identifying attacks with high performance and low costs.