Due to its characteristics including simplicity, lightness, cheap manufacturing costs, and small structure, the Micro Strip Patch Antenna (MPA) plays a significant role in various applications such as aerospace systems, satellites, mobile radar, and other wireless applications. The drawbacks of these current micro strip patch antenna (MPA) design strategies include their low gain, limited frequency bandwidth, and significant return loss. Moreover, the created antenna types are more difficult to build and bigger in size. To solve this issue, the geometrical requirements of the antenna should be improved. The H-shaped antenna for Ku-band applications is designed using the suggested method, which uses a Multilayer Perceptron (MLP) neural network based on the Reptile Search Algorithm (RSA). The fitness value of the RSA is determined using the MLP neural network. The dataset for training the MLP neural network is created using the MATLAB program and includes input parameter values for the substrate height, dielectric constant, length of the patch, resonance frequency, and width as well as output parameter values for gain and return loss. Using the suggested method, the performance of the best-designed antenna is assessed in terms of the radiation pattern, return loss, VSWR, gain, calculation time, directivity, and convergence speed. The suggested technique performs better in experiments and simulations, with 8.89 dB gain, -33.06 dB return loss, and 1.07 VSWR.

Ku-band, Microstrip patch antenna, Multilayer Perceptron Neural Network, and Reptile Search Algorithm.