Mining Trojan horse detection method based on depth canonical correlation analysis

Abstract

The invention relates to the technical field of cloud computing, in particular to a mining Trojan horse detection method based on depth canonical correlation analysis, and has the beneficial effects that the mining Trojan horse detection method based on depth canonical correlation analysis can effectively detect nodes attacked by mining Trojan horses, is beneficial to improving the stability of cloud services, and reduces energy waste at the same time. According to the method, the suspected abnormal node is searched by monitoring the hardware resource use condition and process, and resource occupation under the normal condition can be reduced. Features are extracted by using the neural network in the process of mining Trojan flow detection in flow analysis, so that the situation of flow camouflage can be effectively handled; communication features and behavior features are fused through deep canonical correlation analysis, mutual complementary information is mined from single and one-sided features, and the detection accuracy is improved.

Description

technical field [0001] The invention relates to the technical field of cloud computing, in particular to a mining Trojan detection method based on deep canonical correlation analysis. Background technique [0002] With the continuous advancement of network technology, cloud computing has played an important role in more and more scenarios. Cloud computing provides services through unified scheduling of server resources, and users can access services remotely through the network. [0003] A large number of services run in the cloud service platform. Therefore, the safe and stable operation of the cloud service platform is related to the service quality and user experience. [0004] Aggressive behaviors can affect the operation of cloud service platforms, and mining Trojans are one of them. Mining behavior will greatly increase the resource pressure of the server, resulting in abnormal service; at the same time, it is also an unnecessary waste of energy. In order to reduce t...

AI Technical Summary

Problems solved by technology

[0009] 1. The characteristics of the method of monitoring resource usage are not clear enough, high resource usage is not necessarily occupied by mining Trojan horses, and misjudgments are likely to occur;

[0010] 2. Some mining Trojans can hide the process and are not easy to be found, which affects judgment;

[0011] 3. Judging through traffic analysis requires real-time monitoring of dynamic traffic data and analysis. If the analysis method is complicated, it will affect the efficiency of normal services;

[0012] 4. Part of the traffic is disguised. If only the existing keyword matching rules are used, it cannot be effectively analyzed whether it is mining traffic;

Images