Presentation Information
[WP-B-11]Activation-Function-Dependent Characteristics of Complex-Valued Reservoir-Computing-Based Nonlinear Equalizer for Fiber-Optic Nonlinearity Compensation
○Takumi Yamamoto, Kai Ikuta, Yuta Ito, Tsuyoshi Yamada, Shunya Uchide, Soya Shimomura, Moriya Nakamura (Meiji Univ.)
Keywords:
Artificial intelligence and machine learning for optical transmission systems and subsystems
We investigated the performance of a CVRC-based nonlinear equalizer designed to compensate for fiber-optic nonlinearity when varying the activation function. The performance tended to deteriorate particularly when using an activation function with discontinuous slope changes.