- Biomedical Optics & Medical Imaging
- Defense & Security
- Electronic Imaging & Signal Processing
- Illumination & Displays
- Lasers & Sources
- Micro/Nano Lithography
- Optical Design & Engineering
- Optoelectronics & Communications
- Remote Sensing
- Sensing & Measurement
- Solar & Alternative Energy
- Sign up for Newsroom E-Alerts
- Information for:
Optoelectronics & Communications
IR-to-visible wavelength-band translation in photonic crystal fiber
A parametric process can translate a modulated signal by more than 1μm.
3 August 2006, SPIE Newsroom. DOI: 10.1117/2.1200607.0294
The lack of efficient generation processes has often hindered application of lasers in new wavelength ranges. Each spectral band possesses attributes not available in other spectral regions. For example, mid-IR light can be efficiently transmitted through the atmosphere, while only visible light can propagate any reasonable distance through ocean water.
The entire telecommunication infrastructure is based on nearly lossless transmission of near-IR (1.55μm) light over optical fiber, which motivated the development of a complete infrastructure of sources, modulators, detectors, amplifiers, and instrumentation that work in this spectral region. This technology base was developed over more than three decades and is unlikely ever to be replicated in any other spectral range. Instead of developing new, band-specific technologies, the concept of a universal band translator (UBT) is an appealing alternative that leverages 1.55μm technology across the entire optical spectrum.
Figure 1. The universal band translator (UBT) concept leverages 1.55μm technology across the entire optical spectral range.
Parametric processes are excellent candidates for spectrally-invariant UBT technology extending from visible to IR. For example, the four-photon mixing (FPM) process annihilates two pump photons to amplify the input signal and generate the translated wave at a distant wavelength.1 The translation efficiency generally depends on peak pump power, pump-signal interaction length, and optical nonlinearity of the optical material. In contrast to the case for pulsed lasers, however, for a genuine band translator the peak pump power can not be arbitrarily scaled, because its continuous-wave (CW) operation dictates equal peak and average pump levels. Equally important, pump-signal interaction length is dictated by the operating platform: crystalline materials, while highly nonlinear, can practically be engineered only up to a scale of centimeters.2 An alternative approach, increasing the effective interaction length by enclosing the nonlinear material within an optical cavity,3 dramatically reduces the bandwidth that can be translated. It is therefore not viable for applications requiring high modulation rates, such as communication systems.
Our group is developing a wideband UBT architecture based on single-pass parametric fiber operating from visible to IR. The device combines pump and signal waves in a photonic-crystal fiber (PCF) to produce a widely tunable translated wave. Unlike conventional fiber, which is limited to a standard 1.55μm communication band, PCF can support seamless single-mode transmission from the visible to the IR. More importantly, a precisely engineered PCF transverse structure facilitates the synthesis of the nearly arbitrary dispersion profile required for phase matching between distant signal and pump waves. The latter characteristic is critically important for engineering the long signal-pump interaction lengths needed for efficient band translation.
The optimal translator design relies on the simple notion that, for any band-translation target, a unique PCF dispersion that provides for maximal conversion efficiency can be synthesized. Figure 2(a) illustrates this concept: a signal band (1.55μm) is translated to the visible range (500nm) by choosing a specific PCF dispersion that defines the phase-matching contour. The intersection between the fixed 800nm pump (vertical dashed line) and phase matching contour defines the input (signal) and target (translated) band. The translated (idler) band can then be changed by either pump or signal tuning. More advanced designs that use two instead of one parametric pump can achieve considerable translated powers over a wide spectral range, as illustrated in Figure 2(b): nearly flat spectral response is synthesized in submarine-communication band using a two-pump parametric design.
Figure 2. (a) PCF cross section (inset) defines the phase matching contour used to match signal, pump and translated bands. (b) A two-pump parametric design offers wideband, spectrally equalized response within the translated band.
We have recently performed the first UBT demonstration,4 in which a standard 1.55μm communication signal was modulated and translated to a green wavelength compatible with the submarine transmission window. The demonstration also established an absolute record of 1μm (375THz) for the translation of the modulated signal. PCF with dual zero-dispersion wavelength was selected because of its ability to phase match commercially available (800 to 900nm) pumps, 1.55μm signal and visible translated bands. The experiment used two different PCFs with similar lengths and comparable dispersion characteristics. The translated signal was continuously tuned from 515 to 585nm to demonstrate the available translator bandwidth, in excellent agreement with the estimated PCF phase-matching range of 500–600nm. Figure 3 shows a typical spectrum and the ability to generate fast data carrier in any color. A 155.52MHz pseudorandom bit sequence used to modulate the 1.55μm signal was received error-free in the green band, as indicated by the open eye diagram.
Figure 3. (a) A typical spectrum shows the translation from 1.55μm to 500nm; the wideband operation allows for carrier color tuning (inset). (b) A modulated signal in the 1.55μm band is received in visible (green) band in error-free manner.
In addition to the translation of a single modulated channel, we explored the feasibility of wavelength-division multiplexing (WDM) translation by converting multiple channels from 1.55μm to 500nm. Four channels positioned at 1.541μm, 1.549μm, 1.555μm, and 1.56μm were translated and mapped to 529nm, 530nm, 531nm, and 532nm, thus demonstrating the first WDM generation in the visible band. More importantly, the result points to a natural and simple way for a nearly arbitrary capacity increase in the submarine communication window.
In summary, we have demonstrated the first translation of standard 1.55μm band to the visible spectral range. While the result stands as the absolute translation record to date, it also points to the important potential of parametric PCF technology. In contrast to conventional wavelength converters based on crystalline and cavity-aided platforms, our single-pass approach offers wideband operation in either the continuous-wave or pulsed operating regime. We have used silica PCF to explore 450nm–2.5μm translation regime, and are presently investigating non-silica PCF structures to broaden the spectral coverage to beyond 4μm. Considerable work remains on the understanding of multiple-pump parametric process in dispersion-fluctuating PCF waveguides. We expect that our progress will enable not only spectrally invariant communications, but also allow the use of the superior 1.55μm infrastructure in areas of coherent spectroscopy, imaging and general sensing from visible to infrared.
Photonics Systems Laboratory, UCSD
La Jolla, CA
Prof. Radic received his PhD from The Institute of Optics of University of Rochester in 1995 and was a senior scientist at Corning Research and member of technical staff at Bell Laboratories until 2003. He currently holds the position with the Department of Electrical and Computer Engineering and the California Communication Institute at the University of California. He is an associate editor with Optics Express journal.
1. S. Radic, C. J. McKinstrie, Two-pump fiber parametric amplifiers,
Opt. Fiber Technol.,
Vol: 9, pp. 7-23, 2003.
2. S. E. Bisson, K. M. Armstrong, T. J. Kulp, M. Hartings, Broadly tunable, mode-hop-tuned CW optical parametric oscillator based on periodically poled lithium niobate,
Vol: 40, pp. 6049, 2001.
3. M. Ebrahimzadeh, G. A. Turnbull, T. J. Edwards, D. J. Stothard, I. D. Lindsay, M. H. Dunn, Intracavity CW singly resonant optical parametric oscillators,
Vol: 16, pp. 1499, 1999.