A Study on Intrusion Detection Using Multi-Scale Convolutional Neural Network

With the rapid development of information technology, the protection of network security has become increasingly important, making intrusion detection an indispensable key step in safeguarding networks. In this study, an optimized intrusion detection algorithm based on a multi-scale convolutional neural network is proposed. This algorithm enhances intrusion detection performance by extracting features using convolutional kernels of different sizes. Additionally, the selection process of genetic algorithms is optimized to obtain optimal weights. Through a series of experiments designed during these processes, the superiority of the proposed algorithm over traditional methods is validated, and the impact of different parameters on algorithm performance is analyzed. Experimental results demonstrate a significant improvement in the effectiveness of the enhanced algorithm in intrusion detection tasks.