Low-rate denial of service attack detection method based on Elman neural network

The invention discloses a low-rate denial of service (LDoS) attack detection method based on an Elman neural network, and belongs to the field of network security. The method comprises the following steps: acquiring a data message passing through the key router in the network to form a sample original value; dividing the original value of the sample into a plurality of detection windows at fixed time, detecting by taking the detection windows as units, carrying out original data analysis on data messages in the detection windows, and extracting four characteristic values, namely variance, standard deviation, range and average value, according to fluctuation characteristics and morphological changes of the analyzed data messages; according to the extracted feature values, adding two types of labels to distinguish two types of LDoS attacks and non-LDoS attacks, and adopting an Elman neural network to perform training classification; and inputting to-be-detected data to the trained Elmanneural network for detect