We experimentally demonstrate phase-sensitive amplification in a highly nonlinear and low-dispersion lead-silicate W-type fiber. A phase-sensitive gain variation of 6 dB was observed in a 1.56-m sample of the fiber for a total input pump power of 27.7 dBm.
©2012 Optical Society of America
The increasing demand in global communication traffic fuelled by the rapidly expanding high-bandwidth applications has led to a requirement for more efficient modulation formats than simple on-off keying, which for long has been used as the standard in optical communications. Consequently, the advent of modulation formats, such as differential or binary-phase shift keying (D- or PSK) and differential or quaternary-phase shift keying (D- or QPSK), have led to an increased research interest in phase-sensitive (PS) amplification. Phase-sensitive amplifiers (PSAs) can be used to efficiently amplify signal components of a certain phase, while attenuating any out-of-phase components . PSAs can be implemented by making use of parametric effects in fibers, and have already been shown to allow amplification with a sub 3-dB noise Fig [2,3]. and efficient regeneration of DPSK [4,5] and QPSK signals  by suppressing phase noise.
Efficient parametric processes generally require fiber with a high nonlinearity, low loss, a broadband low dispersion profile and a high stimulated Brillouin scattering (SBS) threshold . Germanium-doped highly nonlinear fibers (HNLFs) have proven to be an important tool for the observation of these effects, mainly due to their excellent dispersion characteristics and low loss. Their application however, requires the use of long lengths of fiber (typically a few hundreds of meters) which can be limiting in terms of the parametric gain bandwidth, device latency and stability. Moreover, SBS eventually limits the amount of narrow linewidth continuous wave pump power that can be coupled into the fiber unless active steps are taken to broaden either the linewidth of the pump  or the SBS gain bandwidth  – both of which add complexity and ultimately compromise the system performance. Nonlinear operation in far shorter fiber lengths can be accommodated through the adoption of soft glasses that exhibit both a far higher nonlinear refractive index and a better SBS figure of merit than silica [10–12]. Among the various soft glasses, lead silicates are commercially available and possess good chemical and thermal properties for fiber fabrication. We have recently shown that an all-solid lead silicate fiber based on a W-type refractive index profile can provide both a high nonlinear coefficient and a wideband low material dispersion profile. We have also demonstrated a uniform four-wave mixing (FWM) conversion efficiency of 0 dB across a 40 nm bandwidth and a high SBS threshold in a 2-m long sample of this fiber . In this paper, we demonstrate PSA operation in this type of fiber. In addition, for the first time we report experiments performed in a fully-fiberized set-up (our previous demonstrations relied on free-space coupling to the lead silicate fiber, greatly limiting the practicality of the applications).
In order to verify that PS operation could be achieved in our fiber, we started our experiments with a single-pump PS parametric fluorescence experiment. Our setup comprised a PSA with the output from a phase-insensitive amplification (PIA) launched into its input. The PIA was based on a 300-m long germanium-doped HNLF with a nonlinear coefficient of 11.6 /W/km, a zero dispersion wavelength (ZDW) of 1553 nm and dispersion slope (DS) of 0.018 ps/nm2/km and was pumped by an EDFA-amplified single CW laser operating at 1565 nm.
A strain gradient was applied to the HNLF during spooling to increase its SBS threshold. The amplified spontaneous emission (ASE) of the EDFA experienced approximately 3 dB of parametric gain at the output of the PIA stage, thus forming a ~30 nm wide phase stabilized signal. Both pump and ASE were subsequently further amplified by another EDFA and were then launched to the fiber under test. The fabrication of the lead-silicate glass W-type index profiled HNLF has been described in detail in . It has a core diameter of 1.66 μm, a length of 1.56 m, a 2.7 dB/m loss, a nonlinear coefficient of 820/W/km, dispersion of ~2 ps/nm/km, and DS of −0.009 ps/nm2/km at 1550 nm. Arc-fusion splicing with an asymmetric configuration [13, 14] was used to splice the lead-silicate glass HNLF with a commercial high NA silica fiber (UHNA3, Nufern Inc.) which itself was then spliced with a conventional SMF28 fiber. 5 dB loss per connection was obtained between the lead silicate fiber and the UHNA3, and 0.25 dB between UHNA3 and SMF28.
Figure 1(a) shows the spectrum measured at the output of the PSA. The wavelength dependent phase mismatch resulting from the dispersive propagation between PIA and the PSA gives rise to the characteristic gain peaks and attenuation troughs in the form of ripples, a clear sign of PS operation . Owing to the low dispersion of the fiber across this whole wavelength range, PS behavior can be observed across the full 30-nm available bandwidth of the parametrically amplified ASE.
We have benchmarked this result against a second sample of soft glass fiber of similar length. This was a bismuth oxide fiber of a similar type to that previously used for an earlier PSA demonstration in . The sample was 2 m long and had a loss of 0.9 dB/m, a nonlinear coefficient of 1100/W/km, and a dispersion of −260 ps/nm/km at 1550 nm. This sample was also spliced to SMF-28 pigtails at both ends and the splice losses were estimated to be approximately 3 dB per splice. Figure 1(b) shows the spectrum obtained at the output of this fiber. The typical PSA ripples are also observed in this case. However, owing to the higher dispersion of this fiber, they decay more rapidly (with wavelength detuning) and are thus restricted to a much narrower spectral region. Therefore, a narrower PSA bandwidth is obtained in the bismuth fiber. In addition, the insets to the two graphs show spectral traces in the vicinity of the pump wavelengths, clearly indicating that for the same input power (approximately 28 dBm) the onset of SBS can be clearly seen in the case of the bismuth fiber, in contrast to the lead-silicate fiber, where even higher pump powers could be used without evidence of SBS (note however, the small difference in length between the two samples).
We next set up a two-pump degenerate PSA in order to measure the phase sensitive gain variation (PSGV) (difference between the PSA maximum gain and maximum de-amplification) achieved with the lead-silicate fiber (Fig. 2 ). The first stage of the setup again consisted of a PIA stage which was configured to ensure phase-locking between the two pumps and the signal . This was done by coupling one pump (pump 1) and the signal from two independent CW lasers into a HNLF (with a length of 300 m, a nonlinear coefficient of 10.5 /W/km and zero-dispersion wavelength at 1550 nm) resulting in the generation of an idler. The weak idler was then used to injection-lock a slave semiconductor laser (pump 2), boosting the power level at this wavelength to that of the pump and the signal. The three optical fields were then fed into a programmable optical filter (Finisar Waveshaper), allowing independent and accurate adjustment of their power levels, to ensure that the powers of the two pumps were equalized throughout the experiment and the signal power remained fixed at a certain level (6 dBm). The pumps and the signal were then amplified with a high power EDFA, and their power levels were monitored on an optical spectrum analyser (OSA) via a tap coupler. The amplified beams were fed through a polarization controller and an isolator at the input of the lead silicate W-type fiber (PSA segment). The signal at the output of the PSA was subsequently filtered using a bandpass filter, removing the two pumps. A piezo-electric transducer (PZT) driven by a 100 Hz signal generator controlled pump 1’s phase prior to PSA. It periodically varied the phase of this field relative to pump 2 and the signal, thereby ultimately causing the PSA to swing between maximum gain and maximum attenuation. We used an electrical time-domain oscilloscope at the output of the system to detect the effect of the PSA swing on the signal (Inset in Fig. 2).
The PSGV was measured as the total input pump power into the fiber was varied from 22.2 dBm to 27.7 dBm. (Note that in estimating these power values at the lead silicate fiber input, we have taken into account a total of 5.3 dB splice loss at the connectorised patchcord). Figure 3(a) (circles) shows the measured PSGV of the lead silicate fiber as a function of the total input pump power. A maximum PSGV of 6 dB was measured for a total input pump power of 27.7 dBm (the maximum we can currently reliably launch from our available amplifiers).
3. Numerical study
The PSGV of the parametric amplifier is often used as a figure of merit for the amplifier’s phase regenerative capability. It determines how the amplifier performs as a phase squeezing device and as a tool for removing phase noise from phase encoded signals. To put the experimentally obtained 6 dB PSGV value into perspective, Fig. 3(b) shows a graph of the PSA phase transfer function for different values of PSGV. As the PSGV value increases from 0 dB, the phase transfer function starts to depart from the linear to the nonlinear regime, where a moderate phase squeezing effect can be observed for a value of 6 dB PSGV.
Moreover, to further understand the potential of this fiber technology, the PSGV of a different lead silicate fiber sample with parameters optimised for the application was simulated numerically. This sample was considered to be identical to the one used here but is 3 m long and has a 2.1 dB/m loss (subsequent fabrication attempts have shown that this fiber loss is readily achievable). The simulation was performed using a commercial modelling tool (VPITransmissionMaker) and followed closely our experimental parameters.
The results of this numerical study are shown in Figs. 3(a) and 3(b) (black curves). The simulations show the margin of improvement in the PSGV of the amplifier achievable using the proposed fiber sample. Figure 3(a) shows that for identical input power levels to those used in the experiment, a value of approximately 18 dB PSGV for an input pump power of 27.7 dBm is achievable using the proposed optimised sample. This value should still be below the SBS threshold for this sample, which we estimate would tolerate at least 33 dBm of total power at the two pump wavelengths. Furthermore, this value of PSGV can be contrasted with the 6 dB value achieved experimentally by comparing the plots of the phase transfer function in Fig. 3(b). Higher PSGV values push the transfer function towards the idealised staircase-shaped function. Indeed for a value of 18 dB PSGV, one could already see the nearly digitised nature of the phase transfer function, where for almost all values of input phase, the output phase is restricted to values close to either 0 or π.
4. Conclusion and future work
We have demonstrated PS amplification in a length of W-type lead silicate fiber. A PS parametric fluorescence experiment demonstrated that PS gain can be observed over a broad bandwidth due to the relatively low dispersion of this fiber enabled by the W-index profile design. This was contrasted to the result achieved with a much higher dispersion bismuth oxide fiber of a similar length, which only showed strong PS gain within a limited bandwidth around the pump wavelength. A two-pump degenerate PSA was employed to measure the PSGV that could be achieved with this fiber, which was measured to be 6 dB for a pump power of 27.7 dBm. There was no need to employ any active SBS suppression schemes in any of the experiments. Furthermore, to gain foresight into the promise of this fiber design, a numerical study was conducted to assess the performance gain achievable using a second fiber sample with better optimised physical characteristics for this application. Using this fiber we believe that a PSGV value in excess of 18 dB should be readily achievable, indicating that such fibers could be employed to implement short-length devices for PSA-based processing applications of wideband signals.
We thank OFS Fitel Denmark for providing the silica HNLFs used. This work has received funding from the European Communities Seventh Framework Programme FP/2007-2013 under grant agreement 224547 (PHASORS) and the EPSRC under grant EP/I01196X: Transforming the Future Internet: The Photonics Hyperhighway.
References and links
1. R. Tang, J. Lasri, P. S. Devgan, V. Grigoryan, P. Kumar, and M. Vasilyev, “Gain characteristics of a frequency nondegenerate phase-sensitive fiber-optic parametric amplifier with phase self-stabilized input,” Opt. Express 13(26), 10483–10493 (2005). [CrossRef] [PubMed]
2. O. Lim, V. Grigoryan, M. Shin, and P. Kumar, “Ultra-low-noise inline fiber-optic phase-sensitive amplifier for analog optical signals,” in Optical Fiber Communication Conference and Exposition and The National Fiber Optic Engineers Conference, OSA Technical Digest (CD) (Optical Society of America, 2008), paper OML3.
3. Z. Tong, C. Lundström, M. Karlsson, M. Vasilyev, and P. A. Andrekson, “Noise performance of a frequency nondegenerate phase-sensitive amplifier with unequalized inputs,” Opt. Lett. 36(5), 722–724 (2011). [CrossRef] [PubMed]
4. R. Slavík, F. Parmigiani, J. Kakande, C. Lundström, M. Sjödin, P. Andrekson, R. Weerasuriya, S. Sygletos, A. D. Ellis, L. Grüner-Nielsen, D. Jakobsen, S. Herstrøm, R. Phelan, J. O'Gorman, A. Bogris, D. Syvridis, S. Dasgupta, P. Petropoulos, and D. J. Richardson, “All-optical phase and amplitude regenerator for next-generation telecommunications systems,” Nat. Photonics 4(10), 690–695 (2010). [CrossRef]
5. K. Croussore and G. F. Li, “Phase regeneration of NRZ-DPSK signals based on symmetric-pump phase-sensitive amplification,” IEEE Photon. Technol. Lett. 19(11), 864–866 (2007). [CrossRef]
6. J. Kakande, A. Bogris, R. Slavík, F. Parmigiani, D. Syvridis, P. Petropoulos, and D. J. Richardson, “First demonstration of all-optical QPSK signal regeneration in a novel multi-format phase sensitive amplifier,” in ECOC (Turin, 2010), paper PDP 3.3.
7. J. Hansryd, P. A. Andrekson, M. Westlund, J. Li, and P. O. Hedekvist, “Fiber-based optical parametric Amplifiers and their Applications,” IEEE J. Sel. Top. Quantum Electron. 8(3), 506–520 (2002). [CrossRef]
8. Y. Aoki, K. Tajima, and I. Mito, “Input power limits of single-mode optical fibers due to stimulated Brillouin scattering in optical communication systems,” J. Lightwave Technol. 6(5), 710–719 (1988). [CrossRef]
9. L. G. Nielsen, S. Dasgupta, M. D. Mermelstein, D. Jakobsen, S. Herstrøm, M. E. V. Pedersen, E. L. Lim, S.-U. Alam, F. Parmigiani, D. J. Richardson, and B. Palsdottir, “A silica based highly nonlinear fiber with improved threshold for stimulated Brillouin scattering,” ECOC (Turin, 2010), paper Tu.4.D.3.
10. H. Ebendorff-Heidepriem, P. Petropoulos, S. Asimakis, V. Finazzi, R. C. Moore, K. Frampton, F. Koizumi, D. J. Richardson, and T. M. Monro, “Bismuth glass holey fibers with high nonlinearity,” Opt. Express 12(21), 5082–5087 (2004). [CrossRef] [PubMed]
11. X. Feng, T. M. Monro, V. Finazzi, R. C. Moore, K. Frampton, P. Petropoulos, and D. J. Richardson, “Extruded single-mode, high nonlinearity, tellurite glass holey fiber,” Electron. Lett. 41(15), 835–837 (2005). [CrossRef]
12. A. Camerlingo, X. Feng, F. Poletti, G. M. Ponzo, F. Parmigiani, P. Horak, M. N. Petrovich, P. Petropoulos, W. H. Loh, and D. J. Richardson, “Near-zero dispersion, highly nonlinear lead-silicate W-type fiber for applications at 1.55 microm,” Opt. Express 18(15), 15747–15756 (2010). [CrossRef] [PubMed]
13. X. Feng, F. Poletti, A. Camerlingo, F. Parmigiani, P. Petropoulos, P. Horak, G. M. Ponzo, M. Petrovich, J. Shi, W. H. Loh, and D. J. Richardson, “Dispersion controlled highly nonlinear fibers for all optical processing at telecoms wavelengths,” Opt. Fiber Technol. 16(6), 378–391 (2010). [CrossRef]
14. X. Feng, J. Shi, G. M. Ponzo, F. Poletti, M. N. Petrovich, N. M. White, P. Petropoulos, M. Ibsen, W. H. Loh, and D. J. Richardson, “Fusion-spliced highly nonlinear soft-glass W-type index profiled fiber with ultra-flattened, low dispersion profile in 1.55μm telecommunication window,” in ECOC (Geneva, 2011), paper We.10.P1.05.
15. J. Kakande, C. Lundström, P. A. Andrekson, Z. Tong, M. Karlsson, P. Petropoulos, F. Parmigiani, and D. J. Richardson, “Detailed characterization of afiber-optic parametric amplifier in phase-sensitive and phase-insensitive operation,” Opt. Express 18(5), 4130–4137 (2010). [CrossRef] [PubMed]
16. K. Croussore and G. F. Li, “Phase and amplitude regeneration of differential phase-shift keyed signals using phase-sensitive amplification,” IEEE J. Sel. Top. Quantum Electron. 14(3), 648–658 (2008). [CrossRef]