skip to main content

Search for: All records

Award ID contains: 1932858

Note: When clicking on a Digital Object Identifier (DOI) number, you will be taken to an external site maintained by the publisher. Some full text articles may not yet be available without a charge during the embargo (administrative interval).
What is a DOI Number?

Some links on this page may take you to non-federal websites. Their policies may differ from this site.

  1. Abstract Solving linear systems, often accomplished by iterative algorithms, is a ubiquitous task in science and engineering. To accommodate the dynamic range and precision requirements, these iterative solvers are carried out on floating-point processing units, which are not efficient in handling large-scale matrix multiplications and inversions. Low-precision, fixed-point digital or analog processors consume only a fraction of the energy per operation than their floating-point counterparts, yet their current usages exclude iterative solvers due to the cumulative computational errors arising from fixed-point arithmetic. In this work, we show that for a simple iterative algorithm, such as Richardson iteration, using a fixed-point processor can provide the same convergence rate and achieve solutions beyond its native precision when combined with residual iteration. These results indicate that power-efficient computing platforms consisting of analog computing devices can be used to solve a broad range of problems without compromising the speed or precision. 
    more » « less
    Free, publicly-accessible full text available December 1, 2024
  2. We propose a coherent multi-dimensional (wavelength, spatial mode, polarization, etc.) photonic tensor accelerator capable of matrix-vector, matrix-matrix, and batch matrix multiplications in a single clock cycle. A proof-of-concept 2x2 matrix-matrix multiplication at 25GBd with 4.67 bit precision was experimentally demonstrated. 
    more » « less
  3. Mode-group multiplexing (MGM) can increase the capacity of short-reach few-mode optical fiber communication links while avoiding complex digital signal processing. In this paper, we present the design and experimental demonstration of a novel mode-group demultiplexer (MG DeMux) using Fabry-Perot (FP) thin-film filters (TFFs). The MG DeMux supports low-crosstalk mode-group demultiplexing, with degeneracies commensurate with those of graded-index (GRIN) multimode fibers. We experimentally demonstrate this functionality by using a commercial six-cavity TFF that was intended for 100 GHz channel spaced wavelength-division multiplexing (WDM) system.

    more » « less
  4. We demonstrate a novel method for finding the coherent transfer matrix (CTM) of a multi-channel transmission medium utilizing backscattering and coherent optical time-domain reflectometry (COTDR). We measured the CTM for two polarizations of a single-mode fiber with ±0.3dB and ±8.5˚ amplitude and phase precisions 
    more » « less
  5. We present a physical neural network (PNN) approach towards multiplane light conversion (MPLC) design. PNN performs a full parameter search with flexible optimization pathways and can tune various design attributes as hyperparameters. 
    more » « less
  6. We propose a coherent multi-dimensional (wavelength, spatial mode, polarization, etc.) photonic tensor accelerator capable of performing high-speed artificial neural network computation. High-speed matrix-vector and matrix-matrix multiplication were experimentally demonstrated. 
    more » « less
  7. null (Ed.)
  8. null (Ed.)
  9. null (Ed.)