What is the Right Theory for Anderson Localization of Light? An Experimental Test

Unsupervised full-color cellular image reconstruction through disordered optical fiber

Recent years have witnessed the tremendous development of fusing fiber-optic imaging with supervised deep learning to enable high-quality imaging of hard-to-reach areas. Nevertheless, the supervised deep learning method imposes strict constraints on fiber-optic imaging systems, where the input objects and the fiber outputs have to be collected in pairs. To unleash the full potential of fiber-optic imaging, unsupervised image reconstruction is in demand. Unfortunately, neither optical fiber bundles nor multimode fibers can achieve a point-to-point transmission of the object with a high sampling density, as is a prerequisite for unsupervised image reconstruction. The recently proposed disordered fibers offer a new solution based on the transverse Anderson localization. Here, we demonstrate unsupervised full-color imaging with a cellular resolution through a meter-long disordered fiber in both transmission and reflection modes. The unsupervised image reconstruction consists of two stages. In the first stage, we perform a pixel-wise standardization on the fiber outputs using the statistics of the objects. In the second stage, we recover the fine details of the reconstructions through a generative adversarial network. Unsupervised image reconstruction does not need paired images, enabling a much more flexible calibration under various conditions. Our new solution achieves full-color high-fidelity cell imaging within a working distance of at least 4 mm by only collecting the fiber outputs after an initial calibration. High imaging robustness is also demonstrated when the disordered fiber is bent with a central angle of 60°. Moreover, the cross-domain generality on unseen objects is shown to be enhanced with a diversified object set.

Details of the topological state transition induced by gradually increased disorder in photonic Chern insulators

Using two well-defined empirical parameters, we numerically investigate the details of the disorder-induced topological state transition (TST) in photonic Chern insulators composed of two-dimensional magnetic photonic crystals (MPCs). The TST undergoes a gradual process, accompanied with some interesting phenomena as the disorder of rod positions in MPCs increases gradually. This kind of TST is determined by the competition among the topologically protected edge state, disorder-induced wave localizations and bulk states in the system. More interestingly, the disorder-induced wave localizations almost have no influence on the one-way propagation of the original photonic topological states (PTSs), and the unidirectional nature of the PTSs at the edge area can survive even when the bulk states arise at stronger disorders. Our results provide detailed demonstrations for the deep understanding of fundamental physics underlying topology and disorder and are also of practical significance in device fabrication with PTSs.

A path to high-quality imaging through disordered optical fibers: a review

In this paper, we review recent progress in disordered optical fiber featuring transverse Anderson localization and its applications for imaging. Anderson localizing optical fiber has a transversely random but longitudinally uniform refractive index profile. The strong scattering from the transversely disordered refractive index profiles generates thousands of guiding modes that are spatially isolated and mainly demonstrate single-mode properties. By making use of these beam transmission channels, robust and high-fidelity imaging transport can be realized. The first disordered optical fiber of this type, the polymer Anderson localizing optical fiber, has been utilized to demonstrate better imaging performance than some of the commercial multicore fibers within a few centimeters transmission distance. To obtain longer transmission lengths and better imaging qualities, glass-air disordered optical fibers are desirable due to their lower loss and larger refractive index contrast. Recently developed high air-filling fraction glass-air disordered fiber can provide bending-independent and high-quality image transport through a meter-long transmission distance. By integrating a deep-learning algorithm with glass-air disordered fiber, a fully flexible, artifact-free, and lensless fiber imaging system is demonstrated, with potential benefits for biomedical and clinical applications. Future research will focus on optimizing structural parameters of disordered optical fiber as well as developing more efficient deep-learning algorithms to further improve the imaging performance.

Dramatic Acceleration of Wave Condensation Mediated by Disorder in Multimode Fibers

Classical nonlinear waves exhibit a phenomenon of condensation that results from the natural irreversible process of thermalization, in analogy with the quantum Bose-Einstein condensation. Wave condensation originates in the divergence of the thermodynamic equilibrium Rayleigh-Jeans distribution, which is responsible for the macroscopic population of the fundamental mode of the system. However, achieving complete thermalization and condensation of incoherent waves through nonlinear optical propagation is known to require prohibitive large interaction lengths. Here, we derive a discrete kinetic equation describing the nonequilibrium evolution of the random wave in the presence of a structural disorder of the medium. Our theory reveals that a weak disorder accelerates the rate of thermalization and condensation by several order of magnitudes. Such a counterintuitive dramatic acceleration of condensation can provide a natural explanation for the recently discovered phenomenon of optical beam self-cleaning. Our experiments in multimode optical fibers report the observation of the transition from an incoherent thermal distribution to wave condensation, with a condensate fraction of up to 60% in the fundamental mode of the waveguide trapping potential.

Mechanical and Electric Control of Photonic Modes in Random Dielectrics

Random dielectrics defines a class of non-absorbing materials where the index of refraction is randomly arranged in space. Whenever the transport mean free path is sufficiently small, light can be confined in modes with very small volume. Random photonic modes have been investigated for their basic physical insights, such as Anderson localization, and recently several applications have been envisioned in the field of renewable energies, telecommunications, and quantum electrodynamics. An advantage for optoelectronics and quantum source integration offered by random systems is their high density of photonic modes, which span a large range of spectral resonances and spatial distributions, thus increasing the probability to match randomly distributed emitters. Conversely, the main disadvantage is the lack of deterministic engineering of one or more of the many random photonic modes achieved. This issue is solved by demonstrating the capability to electrically and mechanically control the random modes at telecom wavelengths in a 2D double membrane system. Very large and reversible mode tuning (up to 50 nm), both toward shorter or longer wavelength, is obtained for random modes with modal volumes of the order of few tens of (λ/n)³ .

Modal area statistics for transverse Anderson localization in disordered optical fibers

We introduce the mode-area probability density function (MA-PDF) as a powerful tool to study transverse Anderson localization (TAL), especially for highly disordered optical fibers. The MA-PDF encompasses all the relevant statistical information on TAL; it relies solely on the physics of the disordered system and is independent of the shape of the external excitation. We explore the scaling of the MA-PDF with the transverse dimensions of the system and show that it converges to a terminal form for structures considerably smaller than those used in experiments, hence substantially reducing the computational cost to study TAL.