Resumen
We use the proposed packet2vec learning algorithm for IDS preprocessing, the basic steps of IDS are as follows. First, the originally collected traffic is split into packets to be truncated into fixed length. Next, the packet2vec learning algorithm is used to obtain local proximity structure features of the packet for preprocessing. Then, the original features of the packet are combined with the local proximity features as the input of deep auto-encoder for IDS. Finally, the accuracy was evaluated with the detection rate in IDS. In addition, the model proposed in this paper can be deployed to the enterprise gateway, dynamically monitor network activities, and connect with the firewall to protect the enterprise?s network from attacks. It can be deployed in a cloud computing environment or a software-defined network to classify traffic, and monitor network behavior and alerts in real time. It can be deployed into a network security situational awareness system for prediction and visualization through spatial feature extraction.