Presentation Information

[8a-N205-5]Deep Neural Network Modeling for Bidirectionally Excited Tamm Plasmon Polariton

〇(D)Naseeb Abdu Taikkaden1, Binu Jose A1, Pranesh Das1, Anirban Sarkar1 (1.NIT Calicut)

Keywords:

Deep learning,Plasmonic devices,Artificial intelligence

The quality of Tamm plasmon polariton (TPP) confinement due to simultaneous bidirectional excitation depends on the wavelength and phase difference of the exciting waves. The abundant combinations of parameters make it tedious to choose the appropriate one for the best confinement quality. Here, a deep neural network (DNN) model for prediction of TPP confinement for different combinations of wavelength and phase difference is developed. The model shows more than 99% accuracy in prediction and can be implemented in various TPP-based devices.