Convolutional Neural Network-based Design of EBG Structures in High-Speed Packages and PCBs

We propose a method of predicting a cutoff frequency band(Stopband) of electromagnetic band gap(EBG) structure that reduces noise using Convolutional Neural Network(CNN). The stopband is predicted by using the structures of ResNet and DenseNet, which are known CNN models. This is compared and analyzed with the results of the existing method.