Enabling antenna design with nano-magnetic materials using machine learning

A machine learning approach to design with magneto dielectric nano-composite (MDNC) substrate for planar inverted-F antenna (PIFA) is presented. A new mixing rule model has been developed. A database of material properties has been created using several particle radius and volume fraction. A second database built with antenna simulations has been developed to complete the machine learning dataset. It is shown that, starting from particle radius and volume fraction of the nano-magnetic material, it is possible to calculate the antenna parameters like gain, bandwidth, radiation efficiency, resonant frequency, and viceversa with good precision by using machine learning techniques.