SIP Flood Attack Detection Method Based on Convolution Neural

Abstract

Aiming at a large number of useless signaling messages generated by Session Initiation Protocol (SIP) in IP networks, the problem of SIP flooding attacks caused by normal call requests cannot be established by users, and a SIP flooding attack detection method based on convolutional neural network is proposed. This method uses a window to perform convolution calculation on the input data, and judges whether there is a SIP message flood attack in the network based on the result of the calculation. Experimental results show that the proposed method can effectively detect SIP flooding attacks in IMS networks.