Optics (physics.optics)

Abstract: Conventional polarizers and polarization beam splitters have a fundamental limit of 50\% efficiency when converting unpolarized light into one specific polarization. Here, we overcome this restriction and achieve near-complete conversion of unpolarized light to a single pure polarization state at several outputs of topology-optimized metasurfaces. Our fabricated metasurface achieves an extinction ratio approaching 100, when characterized with laboratory measurements. We further demonstrate that arbitrary power splitting can be achieved between three or more polarized outputs, offering flexibility in target illumination. Our results provide a path toward greatly improving the efficiency of common unpolarized light sources in a variety of applications requiring pure polarizations.

Abstract: We show how machine learning techniques can be applied for the classification of topological phases in leaky photonic lattices using limited measurement data. We propose an approach based solely on bulk intensity measurements, thus exempt from the need for complicated phase retrieval procedures. In particular, we design a fully connected neural network that accurately determines topological properties from the output intensity distribution in dimerized waveguide arrays with leaky channels, after propagation of a spatially localized initial excitation at a finite distance, in a setting that closely emulates realistic experimental conditions.

Abstract: Guided wave optics, including most prominently fiber optics and integrated photonics, very often considers only one or very few spatial modes of the waveguides. Despite being known and utilized for decades, multi-mode guided wave optics is currently rapidly increasing in parallel with technological improvements and better simulation tools. The physics of multi-mode interactions are usually driven by some initial energy distribution in a number of spatial modes. In this work we introduce how, with free-space input beams having space-time couplings, the different modes can be excited with different complex frequency or time profiles. We cover fundamentals, the coupling with a few simple space-time aberrations, different waveguides, and a number of technical nuances. This concept of space-time initial conditions in multi-mode waveguides will provide yet another tool to study the rich nonlinear interactions in such systems.

Abstract: The widespread adoption of machine learning and other matrix intensive computing algorithms has inspired renewed interest in analog optical computing, which has the potential to perform large-scale matrix multiplications with superior energy scaling and lower latency than digital electronics. However, most existing optical techniques rely on spatial multiplexing to encode and process data in parallel, requiring a large number of high-speed modulators and detectors. More importantly, most of these architectures are restricted to performing a single kernel convolution operation per layer. Here, we introduce a fiber-optic computing architecture based on temporal multiplexing and distributed feedback that performs multiple convolutions on the input data in a single layer (i.e. grouped convolutions). Our approach relies on temporally encoding the input data as an optical pulse train and injecting it into an optical fiber where partial reflectors create a series of delayed copies of the input vector. In this work, we used Rayleigh backscattering in standard single mode fiber as the partial reflectors to encode a series of random kernel transforms. We show that this technique effectively performs a random non-linear projection of the input data into a higher dimensional space which can facilitate a variety of computing tasks, including non-linear principal component analysis, support vector machines, or extreme learning machines. By using a passive fiber to perform the kernel transforms, this approach enables efficient energy scaling with orders of magnitude lower power consumption than GPUs, while using a high-speed modulator and detector maintains low latency and high data-throughput. Finally, our approach is readily integrated with fiber-optic communication links, enabling additional applications such as processing remote sensing data transmitted in the analog domain.

Abstract: Dynamic speckle method is an effective tool for estimation of speed of processes. Speed distribution is encoded in a map built by statistical pointwise processing of time-correlated speckle patterns. For industrial inspection,the outdoor noisy measurement is required. The paper analyzes efficiency of the dynamic speckle method in the presence of environmental noise as phase fluctuations due to lack of vibration isolation and shot noise due to ambient light. Usage of normalized estimates for the case of non-uniform laser illumination is studied. Feasibility of the outdoor measurement has been proven by numerical simulationsof noisy image capture and real experiments with test objects. Good agreement has been demonstrated both in simulation and experiment between the ground truth map and the maps extracted from noisy data.

Abstract: Surface plasmon polaritons and phonon polaritons offer a means of surpassing the diffraction limit of conventional optics and facilitate efficient energy storage, local field enhancement, high sensitivities, benefitting from their subwavelength confinement of light. Unfortunately, losses severely limit the propagation decay length, thus restricting the practical use of polaritons. While optimizing the fabrication technique can help circumvent the scattering loss of imperfect structures, the intrinsic absorption channel leading to heat production cannot be eliminated. Here, we utilize synthetic optical excitation of complex frequency with virtual gain, synthesized by combining the measurements taken at multiple real frequencies, to restore the lossless propagations of phonon polaritons with significantly reduced intrinsic losses. The concept of synthetic complex frequency excitation represents a viable solution to compensate for loss and would benefit applications including photonic circuits, waveguiding and plasmonic/phononic structured illumination microscopy.

Abstract: We report a complete numerical study and supporting experimental results on the spectral characterization of optical aberrations in macroscopic fluidic lenses with tunable focal distance and aperture shape. Using a Shack-Hartmann wave-front sensor we experimentally reconstruct the near-field wave-front transmitted by the fluidic lenses, and we characterize the chromatic aberrations in terms of Zernike polynomials in the visible range. Moreover, we further classify the spectral response of the lenses using clustering techniques, in addition to correlation and convolution measurements. Experimental results are in agreement with our theoretical model of the non-linear deformation of thin ellastic membranes.