Convolutional Neural Network for Coupling Matrix Extraction of Microwave Filters (original) (raw)

Tuning a microwave filter is a challenging problem due to its complexity. Extracting coupling matrix from given S-parameters is essential for ?lter tuning and design. In this paper, a deep-learning-based neural network namely, a convolutional neural network (CNN) is proposed to extract coupling matrix from Sparameters of microwave filters. The training of the proposed CNN is based on a circuit model. In order to exhibit the robustness of the new technique, it is applied on 5-and 8-pole filters and compared with a shallow neural network namely, radial basis function neural network (RBFNN). The results reveal that the CNN can extract the coupling matrix of target S-parameters with high accuracy and speed.