Security Technology of Trusted Internet of Things Based on Machine Learning

The openness of the Internet of Things brings security risks to data transmission. As the most basic and critical technology of the Internet of Things, RFID is the top priority for building a trusted IoT system. The communication between the reader and the electronic tag in the RFID system is non-contact, and the information transmitted on the wireless channel is vulnerable to eavesdropping, tampering and other attacks. In this regard, this paper designs a trusted IoT RFID system based on the artificial neural network model of machine learning and RFID technology. According to the security features of the trusted IoT, combined with the current security requirements, it provides a network for nodes in the network that access the network wirelessly. A model that can ensure the security of IoT information transmission without encryption technology, and uses cross-layer information to perform trust evaluation and detection on IoT nodes to achieve multi-path secure routing. In this paper, the security experiments of the CLMPR protocol and the DSR protocol to maintain the Internet of Things system are carried out, and it is found that the CLMPR protocol can effectively defend against DoS attacks and improve the security of the system.