Soft Computing Approach to Design a Triple-Band Slotted Microstrip Patch Antenna

The design process of antenna structures that meet up-to-date requirements takes a long time and brings a high computational load. In this paper, an approach based on Soft Computing (SC) techniques was used to shorten the design time and to achieve an antenna structure that yields performance characteristics as close as possible to the desired values. In order to obtain a microstrip patch antenna with the targeted characteristics and the best accuracy in a faster way, a Support Vector Machine (SVM)-based regression model was employed. A triple-band microstrip antenna with desired resonance frequencies and gain values was designed by using the Support Vector Regression (SVR) model by introducing multiple slots and arc-truncation to the patch antenna. Simulation results of the High-Frequency Structural Simulator (HFSS) and measurements of implementation of the designed antenna are given. Performance characteristics of the obtained antenna are also compared with those given in the literature, which have triple-band properties. In addition, the antenna was redesigned using the optimization tool in HFSS for comparison. The accuracy of the results and required time for design were compared for both the SVR model approach and the HFSS optimization tool.

Automated summary

This paper uses Soft Computing techniques to shorten the design time of antenna structures and achieve a desired performance. A Support Vector Machine regression model is employed to design a microstrip antenna with targeted characteristics. Comparisons with literature and an optimization tool verify the accuracy and efficiency of this approach.