SDN Attack Identification Model based on CNN Algorithm

As the complexity of network structure increases, so do requirements for architecture are also increasing, and Software Defined Network (SDN) technology has emerged. SDN successfully simplified management, but its open programming nature poses a risk attacks. In complex environments, recognition accuracy traditional models cannot meet speed. view this, this study proposes an attack model based on Convolutional Neural (CNN), aiming to solve problems faced in improve model, ensure security SDN. The used NSL-KDD dataset MIT LL DARPA dataset. performance testing experiment results show that proposed 98.25% recognition, is significantly better than CNN models. reaches 98.25%, superior KNN-PSO model. Verifying superiority further confirming application value identification.