An agricultural network security situation awareness method based on fusion model in digital economy

Nowadays, the digital transformation of agriculture puts forward higher requirements for network security. In response to the complex network traffic data and chaotic attack data cycles in the current agricultural network environment, it is difficult for the network security situation awareness methods to effectively extract network security situation elements and perceive network security status. Therefore, this paper proposes a fusion model-based agricultural network security situation awareness method, namely, MSCNN-ResNeXt-Transformer. Firstly, ResNeXt is improved by fusion model, using Multi-Scale Convolutional Neural Network (MSCNN) instead of a single scale convolution structure. This enables the agricultural network security situation awareness model in digital economy to comprehensively extract network security situational elements from multiple scales. The Efficient Channel Attention (ECA) mechanism is then employed to further refine and characterize the data processed by the improved ResNeXt. Finally, the Transformer is used to optimize the proposed model and improve the accuracy of agricultural network security situation awareness in digital economy. The experimental results show that the accuracy, recall and F1 of MSCNN-ResNeXt-Transformer on MOORE, KDDCUP99 and WSN-DS are significantly better than traditional models, providing effective technical support for agricultural digital security protection.