Research On Key Technologies Of Tactical Wireless Networks Attack Detection

Tactical wireless networks face a serious security situation,as they suffer from various types of attacks and damages,such as resource exhaustion and suppression jamming.It is of great significance to research the attack detection technology of tactical wireless networks.Based on the characteristics of tactical wireless networks and the security threats they face,this paper focuses on the data mining processing model,the attack detection model and algorithm based on network traffic information,and designs an attack detection prototype system to detect the attack behaviors of tactical wireless networks.This paper uses the deep learning method to detect attacks in the monitored data stream of tactical wireless networks.The main contributions of this paper are summarized as follows:1.This paper proposes a attack feature extraction model for the traffic data encrypted by tactical wireless network.According to the attack behavior characteristics of encrypted traffic data,the link layer features and physical layer features are extracted,such as timing correlation and frame size.The model is more suitable for the attack detection algorithm of tactical wireless networks.2.The neural network model ATWI based on CNN and RNN is established.The attack detection algorithm is designed and implemented,which reduces the time and space complexity required for selecting feature data and data processing.Compared with other existing deep learning methods,the algorithm has better classification effect,higher accuracy,lower false detection rate and missed detection rate,with the detection accuracy for AWID four-class dataset reaching 99.16%.3.An attack detection prototype system for tactical wireless networks is designed and implemented without intervening network nodes.The system includes data collection module,data processing module,detection algorithm module and evaluation module.Simulation experiments show that the system can detect typical attacks,including suppression jamming,unfair competition and resource exhaustion.