Beeld op de omslag

Nonlinear optical vector processing using linear silicon photonic circuits for 50 Gb/s memory and string similarity functions

Bewaard in:
Gepubliceerd in:Nature Communications vol. 16, no. 1 (2025), p. 11416-11425
Hoofdauteur: Moschos, T.
Andere auteurs: Pappas, C., Kovaios, S., Roumpos, I., Prapas, A., Tsakyridis, A., Moralis-Pegios, M., Vagionas, C., London, Y., Tossoun, B., Van Vaerenbergh, T., Pleros, N.
Gepubliceerd in:
Nature Publishing Group
Onderwerpen:
Silicon
Insertion loss
Computation
Wave division multiplexing
Germanium
Pattern matching
Microprocessors
Energy efficiency
Architecture
Digital electronics
Computer applications
Circuits
Power consumption
Linear transformations
Machine learning
Optics
Nonlinear phenomena
Wavelength division multiplexing
Electronics
Integrated circuits
Associative memory
Error correction & detection
Nonlinear optics
Wavelength
High speed
Power management
Vector processing (computers)
Photonics
Online toegang:Citation/Abstract
Full Text
Full Text - PDF
Tags: Voeg label toe
Geen labels, Wees de eerste die dit record labelt!
Samenvatting:The need for high-speed, energy-efficient computing in machine learning and communications necessitates innovations beyond conventional electronics to sustain computational power advances without requiring prohibitive energy amounts. Photonics have emerged in various applications, demonstrating significant highlights in optical linear transformations, while if successfully employed can also be used in nonlinear processes and matching functionalities. Towards this we demonstrate nonlinear optical vector processing in the form of hamming distance calculations and content addressable memory banks, using linear silicon photonic circuits at the high-speed of 50 Gb/s, enabling pattern matching and look-up operations. The processor employs a 4×4 crossbar architecture utilizing silicon germanium modulators computing hamming distance between 2-bit optical vectors. It achieves error-rates of ~10⁻³ in binary/ternary content matching, improving state-of-the-art demonstration speeds by >2.5×. Scalability is enhanced through space-wavelength multiplexing via a wavelength-division multiplexing cell, experimentally demonstrated at 50 Gb/s, offering increased computational capacity with reduced insertion losses and power consumption.Photonic technologies enable energy-efficient, high-speed computing, beyond electronics. The work shows nonlinear optical vector processing via linear transformations and wavelength multiplexing, for Hamming Distance and CAMs at 50 Gb/s, via silicon photonics.
ISSN:2041-1723
DOI:10.1038/s41467-025-66286-7
Bron:Health & Medical Collection