Performance Optimization of Multi-Band Microstrip Antennas Using Support Vector Regression for Wireless Communications

Abstract

The increasing demand for compact, high-performance antennas capable of supporting multiple wireless communication standards has driven the development of multiband microstrip antennas. This research presents the design, simulation, and optimization of a multiband microstrip patch antenna operating at 2.4 GHz, 3.5 GHz, and 5.3 GHz, targeting applications in Wi-Fi and WLAN systems. The antenna structure is designed and analyzed using CST Microwave Studio, leveraging its full-wave 3D electromagnetic solver to evaluate key performance metrics including reflection coefficient (S₁₁), gain, bandwidth, and radiation characteristics. To enhance the antenna's performance and reduce the design iteration cycle, support vector regression (SVR), a supervised machine learning technique, is employed. SVR models the nonlinear relationship between the antenna's geometric parameters and its performance outcomes, enabling efficient prediction and optimization. A dataset of 1844 samples is generated through parametric simulations in CST, and the SVR model—using a radial basis function kernel with C = 300, ε = 0.00000000025, and γ = 0.5—is trained to predict return loss and gain across the three target frequencies. The optimized antenna design achieves improved impedance matching, gain enhancement, and bandwidth control at all three frequency bands. Power transfer efficiency exceeds 96% in each band. The results demonstrate that the integration of SVR into the antenna design workflow provides a robust, data-driven approach to achieving multiband performance with high efficiency, making it suitable for next-generation wireless communication systems.

Received: 11 August 2025 | Revised: 15 October 2025 | Accepted: 27 November 2025

Conflicts of Interest

The authors declare that they have no conflicts of interest to this work.

Data Availability Statement

Data are available from the corresponding author upon reasonable request.

Author Contribution Statement

Nejat Abdulwahid Hassen: Conceptualization,  Methodology, Software, Validation, Formal analysis, Investigation, Data curation, Writing - original draft, Visualization, Supervision. Adomeas Asfaw Tafere: Conceptualization, Methodology, Software, Validation, Resources, Data curation,Writing - review & editing, Visualization,Supervision. Murad Ridwan Hassen: Conceptualization, Methodology, Software, Formal analysis, Investigation, Resources, Writing - original draft, Visualization, Supervision, Project administration. Tsega Asresa Mengistu: Conceptualization, Software, Validation, Resources, Writing - review & editing, Visualization, Supervision. Mekete Asmare Huluka: Conceptualization, Software, Validation, Resources, Writing - review & editing, Visualization, Supervision. Amsalu Tessema Adgeh: Conceptualization, Software, Formal analysis, Resources, Writing - review & editing, Visualization.