Ultrabroadband transparent metamaterial absorbers designed by anomalous Brewster effect and gradient impedance matching optimized with deep neural network

Broadband microwave absorbers play a vital role in many applications. This study proposes a metamaterial microwave absorber that combines the anomalous Brewster effect with impedance-gradient matching. The authors explore the broadband response, high absorption, and optical transparency of this structure through electromagnetic simulations and microwave experiments. Additionally, deep-learning techniques are employed to optimize its parameters, leading to further broadening of the operational bandwidth. Such optically transparent microwave absorbers promise wide applications, such as special windows for electromagnetic shielding and stealth.