Network traffic anomaly detection method based on asymmetric convolutional residual memory network

Traffic anomaly detection has become a critical issue in the field of network security due to the ongoing evolution of network assaults. Deep learning-based techniques for detecting network traffic anomalies usually have poor generalization, restricted representation capabilities, and high computational cost. This research suggests an Asymmetric Convolutional Residual Memory Network (ACRMNet) to overcome these issues. The model improves the precision and robustness of anomaly detection by effectively extracting spatial and temporal information from network traffic data by using asymmetric convolution, residual networks, and LSTM modules. Experimental results show that by lowering computing complexity and increasing detection accuracy, ACRMNet performs better than other models, especially when processing complicated traffic data.