Abstract:IoT devices have been widely used in various fields. In order to ensure the security of the Internet of Things and eliminate internal hidden dangers, an automatic security monitoring method for IoT sensing devices is designed based on efficient indexing technology of time series feature data. Combined with the efficient indexing technology of time series feature data to extract the networking information features, on the basis of the message transmission process, distinguish the differences between different traffic data, the characteristics of the system between malicious attack software and sensing devices, calculate the representative values of sample data, and obtain the original information features of the sensing devices of the Internet of Things. Classify the data features, calculate the missing and wrong values in the data, get the screening and optimization results of feature vectors, calculate the training loss function, adjust the threshold value of actual operation, and ensure the accuracy of data feature classification. Build the monitoring model of IoT sensing devices, train discriminators, and conduct automatic monitoring of the Internet of Things. Data packets, bytes and data traffic are identified respectively. This monitoring technology can accurately distinguish benign data and attack data, so as to ensure the security of IoT sensing devices. |