Research on small time scale security intrusion intelligent detection algorithm for dynamic wireless network environment

Wireless network environments are highly dynamic, and attack methods are becoming increasingly sophisticated and intelligent. In order to further improve the detection effectiveness and application performance of wireless network intrusions, this paper focuses on the research of intelligent detection algorithms for wireless network security intrusions at small time scales. The fuzzy C-means clustering algorithm is used to cluster wireless network data at small time scales. Considering the highly dynamic and nonlinear nature of wireless network traffic, sliding window sampling technology is used to extract traffic change feature information. Based on a gradient boosting decision tree, integrate the results of multiple decision trees to achieve intelligent detection of wireless network security intrusion. Experimental results show that the proposed method achieves average sensitivity and specificity of 0.979 and 0.799, respectively, in detecting various types of abnormal intrusions in wireless networks. The accuracy of classifying abnormal traffic is 99.5%, with an error of 0.5%. In the high error rate state of 1 × 10−3, the recall rate of the proposed method can also reach over 85%. The accuracy of intelligent detection is high, and the application effect is good.