Lightweight Internet of Things abnormal traffic detection method based on kernel density estimation

Intrusion detection plays a vital role in computer network security defense, and is one of the key technologies of network security. As the network environment becomes more and more complex, network intrusion behaviors gradually show the characteristics of diversification and intelligence, and it is more and more difficult to be detected. Based on the above reasons, people are concerned about the feasibility and sustainability of existing intrusion detection methods. Specifically, it is difficult for existing intrusion detection algorithms to perfectly abstract the features contained in intrusion behavior, and existing intrusion detection Most of the methods performed poorly on unknown attacks. In response to these problems, the paper proposes an intrusion detection algorithm DAE-3WD based on noise reduction autoencoder and three decision-making. This method uses a denoising autoencoder to extract features from high-dimensional data, uses multiple feature extractions to construct a multi-granular feature spac