Cybersecurity faces constant challenges from increasingly sophisticated network attacks. Recent research shows machine learning can improve attack detection by training models on large labeled datasets. However, obtaining sufficient labeled data is difficult for internal networks. We propose a deep transfer learning model to learn common knowledge from domains with different features and distributions. The model has two feature projection networks to transform heterogeneous features into a common space, and a classification network then predicts transformed features into labels. To align probability distributions for two domains, maximum mean discrepancy (MMD) is used to compute distribution distance alongside classification loss. Though the target domain only has a few labeled samples, unlabeled samples are adequate for computing MMD to align unconditional distributions. In addition, we apply a soft classification scheme on unlabeled data to compute MMD over classes to further align conditional distributions. Experiments between NSL-KDD, UNSW-NB15, and CICIDS2017 validate that the method substantially improves cross-domain network attack detection accuracy.
In order to reach ubiquitous perception in Internet of things, which confusion of network deployment can't generate when information collection, some security problems are considered in IOT, such as key management, privacy protection and data processing. A key management scheme based on Homomorphics Encryption and China Remainder Theorem is proposed, in which a layer network model based on node location is given to deploy network infrastructure. Double key pools are introduced for key management and distribution, which save network overhead and node resource consumption. According to the use of homomorphic encryption for the note privacy information to protect users' privacy and keep the data processing safe. A security network model and pair-key discover probability simulation is given, which shows this scheme have good connectivity and security.