Recognition Method of Abnormal Data in Local Area Network Common Channel Based on Convolutional Neural Network

Abstract

In order to strengthen the accuracy of the recognition of abnormal data in the local area network common channel and improve the recognition efficiency, the convolutional network method is applied to data recognition, and the recognition method of the abnormal data in the local area network common channel based on the convolutional neural network is studied. First, the convolutional neural network structure is analyzed and the three main parts of the convolutional neural network are outlined. Second, the optimization algorithm flow of the abnormal data recognition of the common channel of the local area network is designed, and the specific calculation steps of the data recognition are obtained. Finally, a comparative experiment is designed to compare the traditional recognition method of abnormal data based on data stream and the recognition method of abnormal data based on log analysis with the recognition method of abnormal data in local area network common channel based on convolutional neural network designed in this paper. In this way, the loss value and recognition accuracy data are obtained, and the optimal abnormal data recognition method is judged through the data.