Internet of Things edge lightweight DDoS detection method based on deep learning

The invention provides an Internet of Things edge lightweight DDoS detection method based on deep learning, which comprises the following steps: firstly, carrying out data preprocessing on original traffic generated by Internet of Things edge equipment, and then, carrying out DDoS dichotomy anomaly detection on cut original traffic data by adopting a Bi-GRU bidirectional recurrent neural network; then, according to different module combinations in the Bi-GRU, obtaining composite features of the traffic, and using a multi-classification result of ShuffleNetV2 to optimize DDoS dichotomy anomaly detection at the same time; then, DDoS traffic is intercepted based on an anomaly detection result, and large-scale attack traffic is intercepted at the edge of the network; and finally, through edge deployment, real-time performance and lightweight design of an Internet of Things module, constructing an Internet of Things edge lightweight DDoS detection model based on a deep neural network for lightweight, automatic and