Lightweight Iot Malicious Traffic Awareness Method Based on Multi-Model Collaboration

With the wide application of Internet of Things (IoT) devices, their security problems are becoming increasingly prominent. IoT devices often have limited computing and storage capabilities, making lightweight malicious traffic detection for IoT devices a complex and challenging problem. To solve the above problems, this paper proposes a lightweight IoT malicious traffic awareness method (MTAM-MMC) based on multi-model collaboration. Firstly, a traffic data enhancement algorithm based on synthetic minority oversampling is adopted to solve the problem of data imbalance and alleviate the impact of uneven data distribution on malicious traffic perception performance. Then, a robust feature selection algorithm of model importance evaluation for traffic feature selection is used to select the feature subset that has the greatest impact on malicious traffic detection, reducing the complexity of the model and improving the training efficiency of the model. Finally, combining the malicious traffic detection algorithm based on grid search and random forest, a lightweight malicious traffic perception classifier is constructed, and a large language model is used to realize the dynamic optimization of the multi-model collaborative malicious traffic perception method, so as to increase the interpretability of the perception results and achieve efficient, accurate and robust malicious traffic perception.