Optimization of High-Performance Microstrip Antenna Design Through Machine Learning-Enhanced Parameter Sweeping Techniques

Microstrip antennas (MSAs) have attracted a lot of scholarly interest because of its small size, lightweight design, and easy integration with modern wireless communication systems. However, achieving ideal presentation in terms of bandwidth, improvement, and directivity remains a major challenge. Notably, developments in evolutionary algorithms, machine learning (ML), and parameter sweep optimization have completely changed the way MSAs are designed. Rapid and accurate antenna parameter calibration is now made possible by techniques like parameter sweep (PS), particle swarm optimization (PSO), deep learning (DL), and surrogate modeling. Furthermore, advancements in reconfigurable antennas, graphene-based substrates, and metamaterials have greatly improved the performance of next-generation communication systems, including satellite networks, wireless communication, 5G, 6G, and healthcare systems. This paper proposals a systematic examination of the most recent techniques in MSA Parameter Sweeping optimization, emphasizing their influence on miniaturization, multiband functionality, and enhanced radiation characteristics. The convergence of artificial intelligence and advanced materials is poised to propel forthcoming advancements in adaptive and high-efficiency microstrip antenna designs. Parameter Sweeping optimization in MSA design entails the systematic alteration of critical parameters such as patch dimensions, substrate attributes, and feed positions to evaluate their effects on performance. By progressively modifying one parameter at a time while observing return loss, bandwidth, and radiation efficiency, an optimal design configuration can be realized. This approach elucidates sensitivity and trade-offs, thus assisting in the meticulous adjustment of antenna characteristics. It is frequently employed in conjunction with simulations to expedite the manual design process.