Related Experiment Video
Updated: May 17, 2026

Using Informational Connectivity to Measure the Synchronous Emergence of fMRI Multi-voxel Information Across Time
Published on: July 1, 2014
Real-time spectral interferometry analysis of multi-soliton dynamics based on an unsupervised physical information
None:
The real-time characterization of soliton interaction parameters is of great significance for improving the performance of passively mode-locked fiber lasers and exploring the general laws of dissipative nonlinear systems. Currently, the extraction of soliton dynamic parameters based on deep learning still relies on data labels and lacks physical analyzability. Here, we propose a new, to the best of our knowledge, unsupervised physical information embedding neural network that can unsupervisedly analyze real-time interferometric spectra of multi-soliton, achieving a more dimensional extraction of internal parameters of soliton molecules with a lighter neural network structure. We applied this model to analyze the dynamics of non-stationary double-soliton molecule and triple-soliton molecule, and the PCCs between the reconstructed dispersion Fourier transform spectra and the original dispersion Fourier transform spectra are higher than 0.94 and 0.9, respectively. This model effectively extracts dynamic parameters such as relative separation, relative phase, relative amplitude, and pulse width during soliton interaction, revealing the high-dimensional characteristics of soliton interaction.
Related Concept Videos
IR Spectrum Peak Splitting: Symmetric vs Asymmetric Vibrations
¹H NMR: Interpreting Distorted and Overlapping Signals
As Δν decreases and the signals move closer, the doublets appear increasingly distorted. The intensities of the inner lines increase at the cost of those of the outer lines as the signals are slanted or...
Insensitive Nuclei Enhanced by Polarization Transfer (INEPT)
