AWG-based optical switches performance using crosstalk limiting schedulers

  • Authors:

  • D. FernáNdez Hermida;M. Rodelgo Lacruz;A. Bianco;D. Cuda;G. Gavilanes Castillo;C. LóPez Bravo;F. J. GonzáLez CastañO

  • Affiliations:

  • Grupo de Tecnologías de la Información, Departamento de Ingeniería Telemática, Universidad de Vigo, EI Telecomunicación, Campus, 36310 Vigo, Spain;GRADIANT, Edificio CITEXVI, Campus, 36310 Vigo, Spain;Dipartimento di Elettronica, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Turin, Italy;Dipartimento di Elettronica, Politecnico di Torino, Corso Duca degli Abruzzi 24, 10129 Turin, Italy;Multilayer Wireless Solutions, Istituto Superiore Mario Boella, Via P.C. Boggio, 61, 10100 TORINO, Italia;Grupo de Tecnologías de la Información, Departamento de Ingeniería Telemática, Universidad de Vigo, EI Telecomunicación, Campus, 36310 Vigo, Spain;Grupo de Tecnologías de la Información, Departamento de Ingeniería Telemática, Universidad de Vigo, EI Telecomunicación, Campus, 36310 Vigo, Spain and GRADIANT, Edificio C ...

  • Venue:
  • Computer Networks: The International Journal of Computer and Telecommunications Networking
  • Year:
  • 2012

Quantified Score

Hi-index 0.00

Visualization

Abstract

According to a generalized opinion of the scientific community, Arrayed Waveguide Gratings (AWG) are one of the most promising passive optical devices for terabit optical switching systems. These devices permit to build high-performance all-optical WDM switches thanks to their wavelength routing capabilities, in addition to their high information density and lower power consumption. In previous works, we showed that wavelength reuse across different ports introduces in-band crosstalk which strongly limits scalability of AWG-based backplanes. We also proved that this limitation can be overcome by modified scheduling algorithms that reduce the probability of reusing the same wavelength in different ports of the AWG device, significantly reducing or even avoiding the effect of in-band crosstalk. In this paper, we extend several previously proposed scheduling algorithms to enhance their performance. The new algorithms permits to build AWG-based switches of larger sizes while maintaining small bit error rates (BER).