Dimension optimization of multi-band microstrip antennas using deep learning methods

The electromagnetic frequency spectrum is divided into different sub-frequency bands. These bands are allocated for applications. In these days, devices operating in multiple provide significant advantages. Devices require antenna structures to operate Microstrip antennas have become prominent with their small size, portable and easy integration other systems. this study, microstrip structure which can work multi designed. At the same time, it was used deep learning methods optimization of sizes ensure designed a shorter time. frequencies C X band as seen obtained results. According IEEE standards, determined between 4 GHz 8 GHz; 12 range. proposed structure, ability multi-band achieved by means C-shaped array. that will be process, five Long Short Term Memory (LSTM) models used. most important advantage achieve satisfactory results identifying necessary features solving difficult time consuming problems its own ability. context, 52 pieces data were produced. 40 training process test stage. lowest root mean square error (RMSE) performance LSTM-1 + Dropout layer-1 LSTM -2 layer-2 1.0161 value. method evaluated High Frequency Simulation Software (HFSS) program. experimental results, observed produced model very close each other.