Solving the Nonlinear Schrödinger Equation in Optical Fibers Using Physics-informed Neural Network
Xiaotian Jiang, Danshi Wang, Qirui Fan, Min Zhang, Chao Lü, Alan Pak Tao Lau
Abstract
We constructed a physics-informed neural network (PINN) to solve the nonlinear Schrödinger equation for different input waveforms. Results show that PINN can accurately characterize pulse evolution in fibers with less complexity to SSFM methods.
Topics & Concepts
Artificial neural networkNonlinear systemPulse (music)WaveformPhysicsNonlinear Schrödinger equationNonlinear opticsOptical fiberSchrödinger equationStatistical physicsApplied mathematicsComputer scienceMathematicsOpticsQuantum mechanicsArtificial intelligenceDetectorVoltageModel Reduction and Neural NetworksAdvanced Fiber Laser TechnologiesNeural Networks and Reservoir Computing