With the rapid development of the internet of things (IoT) technology, IoT devices have brought great convenience to agriculture, industry, and our daily lives. However, there exist numerous vulnerabilities and lack of efficient protective measures in IoT, thus its devices can be easily infected by malware. It is of great importance to improve the accuracy of malware classification for detecting and preventing the IoT malware. In this article, we adopt the mixture of experts (MoE) neural network to analyze and classify the family of the IoT malware. A classification framework is proposed based on the MoE neural network which utilizes the multitask learning approach and is designed to train multiple neural networks, each of which is responsible for a set of data and tasks. The proposed framework contains three neural networks which are designed to analyze and classify the malware and benignware samples. Especially, an improved hierarchical softmax algorithm based on the MoE neural network is used to distinguish the malware from benignware and get its exact classification. Experiment results show that the presented MoE neural network model can effectively distinguish the malware and benignware. Besides, the MoE neural network is also effective for malware family classification. The comparison of experiment results with other literatures shows that the proposed classification method has better performance.

No abstract

No abstract

No abstract

Internet of Things (IoT) is fast growing. Non-personal computer devices under the umbrella of IoT have been increasingly applied in various fields and will soon account for a significant share of total Internet traffic. However, the security and privacy of IoT and its devices have been challenged by malware, particularly polymorphic worms that rapidly selfpropagate once being launched and vary their appearance over each infection to escape from the detection of signature-based intrusion detection systems. It is well recognized that polymorphic worms are one of the most intrusive threats to IoT security.To build an effective, strong defense for IoT networks against polymorphic worms, this study proposes a machine intelligent system, termed Gram-Restricted Boltzmann Machine (Gram-RBM), which automatically generates generic fingerprints/signatures for the polymorphic worm. Two augmented N-gram-based methods are designed and applied in the derivation of polymorphic worm sequences, also known as fingerprints/signatures. These derived sequences are then optimized using the Gaussian-Bernoulli RBM dimension-reduction algorithm. The results, gained