Hierarchical Network Attack Identification and Unknown Attack Detection Method Based on Deep Learning

Abstract

The invention discloses a layered network attack identification and unknown attack detection method based on deep learning, which includes: using an autoencoder to learn the behavior patterns of normal traffic and abnormal traffic, so as to use the learned autoencoder to distinguish the untested The traffic is normal traffic or abnormal traffic; the deep neural network is used as the discriminant model to determine whether the attack type of the traffic to be tested is a known attack category or an unknown category; the results of the fusion of the autoencoder and the discriminant model are used to complete the network attack Classification and unknown attack detection. This method can not only identify normal traffic and classify known abnormal traffic, but also detect new and unknown network attacks, combining the respective advantages of generative model and discriminant model to improve detection accuracy.

Description

technical field [0001] The invention relates to the technical field of computer networks, in particular to a layered network attack identification and unknown attack detection method based on deep learning. Background technique [0002] With the rapid development of network technology, the Internet has covered all aspects of social life and has had a profound impact on social life. Although the wide application of the network has promoted the rapid development of the social economy, it has also become an important medium for hackers and criminals to spread malicious software and carry out network attacks. In recent years, new network attack methods targeting network protocol and application vulnerabilities have emerged in an endless stream. Network security detection has become one of the important tasks in network supervision and operation and maintenance, especially the detection of unknown zero-day network attacks has attracted widespread attention. [0003] Traditional ...

AI Technical Summary

Problems solved by technology

Anomaly detection only needs normal traffic as input data, and the data requirements are small, and to a certain extent, unknown intrusion behaviors can be detected. problems of classification

Images