An integrated method of behavior baseline self-learning and intrusion detection for large scale IoT terminals

Abstract

The invention discloses a large-scale IOT terminal behavior baseline self-learning and intrusion detection integration method, which relates to the IOT security technology field.This method realizes the efficient collection and preprocessing of multi-dimensional behavior data of large-scale IOT terminals by constructing a hierarchical data collection architecture;Based on the improved incremental clustering algorithm, the self-learning construction of behavior baseline is completed, and the behavior pattern drift perception mechanism is introduced synchronously to realize the dynamic update of baseline;A two-dimensional intrusion detection model integrating baseline deviation and abnormal behavior is designed, and the accurate identification of intrusion behavior is achieved by combining terminal attribute association analysis;At the same time, the linkage optimization mechanism- ofself learning and intrusion detection is established to improve the adaptive ability and detection efficiency of the whole system.