Perancangan Antena Mikrostrip Patch Rectangular Array 3x1 dengan Slot Rectangular pada Frekuensi 2,45 Ghz Berbasis Random Forest

Abstract

The advancement of wireless communication technology demands devices that can support efficient data transmission, one of which is the microstrip antenna. This type of antenna is known for its compact size, lightweight structure, and ease of integration with modern systems. This study designs and optimizes a 3×1 rectangular patch microstrip antenna array with a rectangular slot operating at 2.45 GHz. The optimization process is carried out using the Random Forest machine learning algorithm to achieve more accurate and efficient antenna performance. Simulation results show that the designed antenna achieved a VSWR of 1.08, return loss of –28.72 dB, gain of 6.11 dBi, and bandwidth of 52 MHz. The Random Forest model provided accurate predictions with MSE of 5.1643 and MAE of 1.3021. Fabrication and testing using a VNA confirmed that the proposed antenna operates effectively at 2.45 GHz and supports communication in the ISM band. The machine learning–based approach proved effective in providing optimal antenna design recommendations, and this research can serve as a reference for the development of smart antennas and as learning material at the Intelligent Radio Systems Laboratory, Universitas Sumatera Utara.