SPIE Digital Library Get updates from SPIE Newsroom
  • Newsroom Home
  • Astronomy
  • Biomedical Optics & Medical Imaging
  • Defense & Security
  • Electronic Imaging & Signal Processing
  • Illumination & Displays
  • Lasers & Sources
  • Micro/Nano Lithography
  • Nanotechnology
  • Optical Design & Engineering
  • Optoelectronics & Communications
  • Remote Sensing
  • Sensing & Measurement
  • Solar & Alternative Energy
  • Sign up for Newsroom E-Alerts
  • Information for:
    Advertisers
SPIE Defense + Commercial Sensing 2017 | Call for Papers

Journal of Medical Imaging | Learn more

SPIE PRESS




Print PageEmail PageView PDF

Optoelectronics & Communications

Ultrafast operations in all-optical devices for communications networks

Future performance of light-based networks will depend on improved transmission bandwidth to avoid bottlenecks.
14 January 2008, SPIE Newsroom. DOI: 10.1117/2.1200801.0775

Each channel in today's high-performance wavelength division multiple access (WDMA) optical networks may run up to 40Gbit/s. Future targets promise more than 100Gbit/s for aggregate throughput of several terabits per second over a single optical fiber. Devices for carrying out wavelength conversion and data regeneration will require transmission bandwidth that state-of-the-art optoelectronics, limited to few hundred Gbit/s,1 cannot support. However, recent advances can avoid this potential bottleneck. Novel all-optical devices will deliver improved bandwidth for ultrafast switching, wavelength conversion, and so-called 3R (reamplification, reshaping, retiming) regeneration.

We have developed and demonstrated an all-optical device capable of such ultrafast operation. The terahertz optical asymmetric demultiplexer (TOAD)2 is based on a nonlinear interferometer that has been used in various applications for all-optical data processing.2–4,5 The Sagnac version (see Figure 1) uses an intraloop 2(×)2 coupler to inject control pulses to the semiconductor optical amplifier (SOA) employed as the nonlinear element. Input data splits clockwise and counterclockwise to counterpropagate around the loop and reach the SOA at slightly different times. The control pulse arrives just before the counterclockwise component of the data bit (to be gated), but just after its clockwise component, thus inducing nonlinearities. This causes the two components to experience different phase shifts and, consequently, to recombine and exit the loop at the output port. All other data signals, for which the counterclockwise components do not straddle the control pulse's arrival at the SOA, exit the loop at the input port. The following equation describes the transfer function at the TOAD-switching window:


Figure 1. Schematic diagram of the terahertz optical asymmetric demultiplexer (TOAD). Δ x: Offset that defines the width of the switching window. SOA: Semiconductor optical amplifier.

Gcw, Gccw, φcw, and φccw are the gain and phase experienced by the clockwise and counterclockwise propagating signals, respectively. The offset Δ x defines the width of the switching window Δ τ = 2Δ xsoa/cfiber . If Δ τ is set small (say a few picoseconds), one can obtain all-optical processing at speeds approaching terabits per second.

TOAD-based data processing has already provided highly promising results. Kailight Photonics reported all-optical 2R (retime and retransmit) on a single wavelength-division multiplexing (WMD) channel.3 Up to 17ps of timing jitter from the OC192 (9.6Gbit/s) signal was all-optically eliminated with the TOAD-based regenerator.4 Our approach to all-optical 3R regeneration relies on a patented all-optical TOAD-based sampling technology.

The concept is shown schematically in Figure 2. Input to the regenerator consists of four 10Gbit/s WDM non-return-to-zero (NRZ) data signals that are impaired by noise and timing jitter. First, dispersion-compensating fiber is used to eliminate transmission dispersion. Then TOAD-1 performs bit-parallel gating during which four WDM channels are simultaneously gated, thus converting NRZ WDM data into a perfectly retimed return to zero (RZ). The next stage (A–B) performs bit-interleaving of the retimed RZ data. This stage uses a 1(×)4 thin-film filter demux (delay lines) 4(×)1 power combiner structure to interleave, or ‘stagger,’ the four WDM channels from the 10–40Gbit/s data stream. Its output (B) is then amplified by EDFA (erbium-doped fiber amplifier) and fed into the control port of TOAD-2. The input port of TOAD-2 is fed with multiwavelength picosecond pulses generated by spectrally sliced optical supercontinuum having minimal noise and low timing jitter (<25fs). TOAD−2 serves as an ultrafast all-optical modulator. It uses interleaved WDM data at its control port as the time-gate control for clean multiwavelength pulses at its input port, thus simultaneously reshaping all four WDM channels. At the output port (C) retimed, reshaped, and reamplified data will exit, and 3R regeneration is achieved.


Figure 2. TOAD-based wavelength-division multiplexing (WDM) multichannel 3R regenerator. NRZ: Non-return to zero. DCF: Dispersion compensation fiber. EDFA: Erbium-doped fiber amplifier. DeMUX: Demultiplexer. MUX: Multiplexer. TFF: Thin-film filter.

We believe that ultrafast all-optical gates will be used in conjunction with high-speed electronics to enhance network capacity. The TOAD has proved a serious candidate and has been extensively studied in various all-optical signal-processing applications with a view to overcoming the developing electronic bottleneck in optical networks.


Ivan Glesk, Paul Prucnal 
Electrical Engineering Department
Princeton University
Princeton, NJ

Ivan Glesk is professor of physics in the Department of Experimental Physics, Comenius University, Bratislava, Slovakia. Currently, he is senior research scholar and manager of the Lightwave Communication Research Laboratory at Princeton University. His research interests encompass all-optical switching, interconnects, and fiber lasers. He is author or coauthor of 200 publications, including 16 book chapters.