Analysis and design of transition radiation in layered uniaxial crystals using tandem neural networks

With the flourishing development of nanophotonics, Cherenkov radiation pattern can be designed to achieve superior performance in particle detection by fine-tuning properties metamaterials such as photonic crystals (PCs) surrounding swift particle. However, sensitive geometry and material PCs, periodicity, unit thickness, dielectric fraction, making direct analysis inverse design difficult. In this article, we propose a systematic method analyze PC-based transition radiation, which is assisted deep learning neural networks. By matching boundary conditions at interfaces, Cherenkov-like multilayered structures resolved analytically using cascading scattering matrix method, despite optical axes not being aligned with electron trajectory. Once well trained, forward networks utilized predict without further electromagnetic simulations; moreover, Tandem have been proposed inversely and/or for desired pattern. Our proposal demonstrates promising strategy dealing layered-medium-based detectors, it extended other emerging metamaterials, time crystals.