Deep learning characterizes optical pulses using speckle patterns at the end of multimode fibers
Deep learning characterizes optical pulses using speckle patterns at the end of multimode fibers lead image
Multimode fibers have been used widely in communication but less attention has been paid to their potential for gathering information of light fields such as characterizing ultrashort optical pulses, which is a difficult task. Xiong et al. demonstrate a technique that allows for single-shot pulse characterization in a simple experimental set-up using a deep learning algorithm for phase recovery.
The artificial neural networks analyzed the nonlinear speckle patterns formed at the distal end of the multimode fiber in order to determine the spectral phase of an unknown pulse.
“I was most surprised by the fact that the artificial neural network could distinguish noise from the information about optical pulses that is encoded in the speckle patterns,” said author Hui Cao.
Deep learning networks have been employed earlier for imaging through multimode fibers but the large amount of data required for training the network was an obstacle. The researchers used the experimentally measured transmission matrix of the multimode fiber to calculate two-photon patterns for the purpose of training the network. This hybrid model made it straightforward to generate numerical data for the specific fiber of the experiment.
“Optic fibers are already used for remote sensing in applications like endoscopy,” Cao said. “We are now discovering multimode fibers are really versatile and when combined with machine learning can be used broadly in spectroscopy, sensing and imaging applications.”
By manipulating the diverse degrees of freedom, multimode fibers can operate as a diffraction-limited microscope, a high-resolution spectrometer, a radio frequency wave-sensor, an optical pulse shaper and a reconfigurable waveplate.
Source: “Deep learning of ultrafast pulses with a multimode fiber,” by Wen Xiong, Brandon Redding, Shai Gertler, Yaron Bromberg, Hemant D. Tagare, and Hui Cao, APL Photonics (2020). The article can be accessed at https://doi.org/10.1063/5.0007037 .
