A Wideband Butterfly Antenna Based on Deep Learning Parameter Optimization Algorithm

A wideband butterfly antenna based on deep learning parameter optimization algorithm is proposed in this paper. Two neural networks are designed to model the mapping relationship of structure parameters to frequency response and frequency response to structure parameters respectively. The parameter optimization algorithm proposed consists of two stages: training NN1 and optimizing parameters using NN2. A wideband butterfly antenna is designed to verify the algorithm. The experiment shows that the structure parameter optimization algorithm proposed can quickly optimize the structure parameters of antenna with the best preformance and save a lot of manpower and time cost.