Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Extinction ratio improvement by pump-modulated four-wave mixing in a dispersion-flattened nonlinear photonic crystal fiber

Open Access Open Access

Abstract

We demonstrate extinction ratio improvement by using pump-modulated four-wave mixing in a dispersion-flattened nonlinear photonic crystal fiber. A 6-dB improvement in the extinction ratio of a degraded return-to-zero signal has been achieved. A power penalty improvement of 3 dB at 10-9 bit-error-rate level is obtained in the 10 Gb/s bit-error-rate measurements.

©2005 Optical Society of America

1. Introduction

All-optical regeneration of degraded signals is desirable for future large-scale optical networks. Among various methods, four-wave mixing (FWM) is a promising technique owing to its high-speed performance and the capability to maintain the same logical polarity. Much research has been focused on the regenerative properties of FWM in dispersion-shifted fibers (DSF) [1–4]. However, these schemes always need a long segment of fiber and the pump wavelength should be placed near the zero dispersion point of the DSF to maintain the phase-matching condition of FWM. The adoption of high-nonlinear (NL) DSF can effectively shorten the required fiber length to about 1 km [5, 6], but the dispersion of the fiber still limits the flexibility of operating wavelength. Recently, our group has demonstrated widely wavelength tunable FWM using a dispersion-flattened nonlinear photonic crystal fiber (PCF) [7]. In this paper, we further demonstrate the extinction ratio improvement properties of FWM in our PCF. With the use of pump-modulated FWM, a degraded 10 Gb/s signal shows improvement of 6 dB in extinction ratio (ER), and a power penalty improvement of 3 dB is measured at the 10-9 BER level. Due to the dispersion-flattened characteristics of the PCF, the setup can operate over 40 nm in the 1550 nm wavelength range.

2. Principle and Experiment

Figure 1 shows the principle of our scheme. For the conventional FWM with ω1 and ω2 launch into a fiber, a new wave ωc=2ω1–ω2 will be generated and its power Pc can be expressed as [2]:

Pc(L)=ηγ2P12P2exp(αL){[1exp(αL)]2α2}

where L is the length of fiber, η is the FWM efficiency of ω3, γ is the nonlinear coefficient of the fiber, and α is the attenuation coefficient of the fiber. Note that Pc is proportional to the square of P1, which acts as the pump of FWM. In our scheme, the degraded input signal with a low extinction ratio is set to ω1 where ω2 is a continuous wave (CW) source. When ω1 is in the “one” state, FWM occurs and ωc is generated as shown in Fig. 1(a). However, when ω1 is in the “zero” state with some residue power due to the low extinction ratio, the main generated signal will be ω3 as shown in Fig. 1(b). The static power transfer function of this scheme will become Pc proportional to P12, so that the unwanted power of ω1 during the “zero” state will be suppressed in ωc and therefore the extinction ratio of the signal can be enhanced.

 figure: Fig. 1.

Fig. 1. Schematic illustration of pump-modulated four-wave mixing for extinct ratio enhancement. (a) The degraded input signal ω1 is in “one” state with ω2 as the CW source and ωc as the converted signal. (b) ω1 is in “zero” state with ωc suppressed.

Download Full Size | PDF

 figure: Fig. 2.

Fig. 2. Experimental setup on pump-modulated four-wave mixing in a nonlinear photonic crystal fiber (PCF). The inset shows the microscopic picture of the mircrostructured region of the PCF. EDFA: erbium-doped fiber amplifier; PC: polarization controller; FFP: fiber Fabry-Perot filter.

Download Full Size | PDF

Figure 2 shows our experimental setup on pump-modulated FWM in PCF. The input signal (P1) is a 231-1 bits pseudorandom return-to-zero (RZ) signal at 10 Gb/s. It is prepared by external modulation of a 10 GHz gain-switched distributed-feedback (DFB) laser. The ER of the RZ signal is intentionally set to 6 dB and the average power is 0 dBm. Then the input signal is combined with a wavelength tunable continuous-wave (CW) pump (P2) using a 3-dB coupler. The spectral width of P2 is about 1 GHz and no stimulated Brillouin back scattering from the fiber is observed with a pump power of 300 mW in a preliminary experiment. In order to demonstrate the extinction ratio improvement of FWM, the input signal P1 will act as the pump of FWM, which is different from conventional FWM [2]. Then the combined signal is amplified by an erbium-doped fiber amplifier (EDFA) with a saturation power of 25 dBm and launched into a segment of 64 m long PCF. The PCF is used as the nonlinear medium for FWM of the input signal and pump, resulting in generation of the converted signal. A tunable fiber Fabry-Perot filter (FFP) with a 3-dB bandwidth of 1 nm is used to filter out the converted signal.

The inset of Fig. 2 shows the microscopic picture of the mircrostructured region of the PCF, which is supplied by Crystal Fibre A/S. The fiber has a three-fold symmetric hybrid core region with a core diameter of 1.5 μm. It is worth noting that the overall dispersion of this PCF is flat over a wide wavelength range (less than -3 ps/(km·nm) over 1500–1600 nm) with a nonlinearity coefficient of 11.2 (W·km)-1 [8]. The dispersion variation is less than 1 ps/(km·nm) in the range of 1465–1655 nm, with a dispersion slope of less than 1×10-3 ps/(km·nm2) in the 1550 nm range. Also, the attenuation of the fiber is less than 10 dB/km in the 1550 nm range and both ends are spliced to standard single-mode fiber, yielding a total loss of 2.6 dB.

 figure: Fig. 3.

Fig. 3. Optical spectra measured at different positions of the setup.

Download Full Size | PDF

3. Results and Discussion

Figure 3 shows the optical spectra at different positions of the experimental setup, which is measured by an optical spectrum analyzer with a resolution of 0.08 nm. The dotted curve is the spectrum of the input signal (P1) with a peak wavelength at 1550.0 nm. The dash curve shows the spectrum after the PCF without filtering with P2 at 1555.8 nm. Note that FWM occurs and the converted signal is generated in the shorter wavelength side with an optical signal to noise ratio of 20 dB. Also, the spectral broadening of P1 and P2 is due to self-phase modulation and cross-phase modulation. The peak power of P1, P2 and PC are 2.9 dBm, 3.5 dBm and -17.2 dBm, respectively. Finally, we use a tunable FFP to filter out the converted signal. The solid curve shows the filtered converted output signal (C) from the FWM spectrum, with a peak wavelength of 1544.2 nm. Fig. 4(a) and (b) show the eye-diagrams of the degraded 10 Gb/s input RZ signal and the corresponding filtered output, respectively. Note that the ER of the input signal is degraded to 6 dB. The eye-diagram in Fig. 4(b) shows that the signal is restored and the ER is improved to 12 dB with a widely-open eye. In order to investigate the system performance of our setup, a 10 Gb/s BER measurement is performed. Fig. 5 plots the output BER against the received optical power. The result shows the BER characteristics of the degraded signal and the output signal. Considering the 10-9 BER level of the output signal, a power penalty improvement of 3 dB is obtained compared with the degraded signal. Note that the static power transfer function of this setup is Pc proportional to P12, noise in the signal “one” state is enhanced. Thus this scheme cannot be applied to signal that suffers from signal-spontaneous beat noise due to optical amplifiers.

 figure: Fig. 4.

Fig. 4. Eye diagrams of (a) degraded 10 Gb/s RZ input signal and (b) the corresponding output.

Download Full Size | PDF

 figure: Fig. 5.

Fig. 5. Plot of the bit-error-rate against the received optical power in a 10 Gb/s BER measurement.

Download Full Size | PDF

It is worth noting that since the dispersion of the PCF in the experiment is relatively small and flat over a wide wavelength range (1465–1655 nm), our setup can operate over a wider wavelength range compared with that using a normal DSF as the nonlinear medium. Fig. 6 shows the power of the output signal as well as ER improvement with different output wavelength. The input wavelength P1 (1550.0 nm) is the same as that shown in Fig. 3. A 3-dB tuning range over 40 nm is obtained and the results show that ER improvements of 5 to 6 dB are achieved. Note that the dispersion of the PCF is flattened over the tuning range. Although the fiber has a finite dispersion of about -3 ps/(km·nm) in this region, different wavelengths will face similar dispersion and phase matching can be maintained.

Also, previous report shows that when FWM operates in anomalous regime of the fiber, the filtered output signal often becomes noisy due to unstable excitation of random higher order solitons [9]. However, in our setup the PCF has a normal dispersion over the 1550 nm window, thus leading to a wider operation wavelength range. Since the FWM process is polarization sensitive, one can adopt the polarization diversity scheme to solve this problem [7, 10]. Since the structure of the PCF used in our experiment has a three-fold symmetry, there is very low inherent birefringence in the design [8] and therefore similar FWM effect is expected for both clockwise and counterclockwise propagating signals with orthogonal polarizations in the loop. Based on our previous work [7], we believe that the polarization sensitivity can be limited to be less than 0.3 dB. Note that without the scheme of polarization diversity, more than 10 dB variation in the output power has been experimentally observed.

 figure: Fig. 6.

Fig. 6. Plot of the output power and extinction ratio improvement against different output wavelength.

Download Full Size | PDF

4. Conclusion

Widely wavelength tunable extinction ratio improvement properties of pump-modulated four-wave mixing have been demonstrated using a nonlinear photonic crystal fiber. The output signal shows a 6 dB extinction ratio improvement with a power penalty improvement of 3 dB at 10-9 BER level in a 10 Gb/s BER measurement. With the dispersion flattened PCF, an operation range over 40 nm is obtained.

Acknowledgement

The authors would like to acknowledge Crystal Fibre A/S Denmark for providing the photonic crystal fiber. The authors would also like to thank S. H. Lee for her assistance in performing the experiment. The work described in this paper was supported by the Research Grants Council of the HKSAR, China (Project Nos. CUHK 4196/03E, CUHK 4184/04E).

References and links

1 . A. Argyris , H. Simos , A. Ikiades , E. Roditi , and D. Syvridis , “ Extinction ratio improvement by four-wave mixing in dispersion-shifted fibre ,” Electon. Lett. 39 , 230 – 232 ( 2003 ). [CrossRef]  

2 . A. Bogris and D. Syvridis , “ Regenerative properties of a pump-modulated four-wave mixing scheme in dispersion-shifted fibers ,” J. Lightwave Technol. 21 , 1892 – 1920 ( 2003 ). [CrossRef]  

3 . E. Ciaramella and S. Trillo , All-optical signal reshaping via four-wave mixing in optical fibers , IEEE Photon. Technol. Lett. 12 , 849 – 851 ( 2000 ). [CrossRef]  

4 . E. Ciaramella , F. Curti , and S. Trillo , All-optical signal reshaping by means of four-wave mixing in optical fibers , IEEE Photon. Technol. Lett. 13 , 142 – 144 ( 2001 ). [CrossRef]  

5 . S. Radic , C. J. McKinstrie , R. M. Jopson , J. C. Centanni , and A. R. Chraplyvy , “ All-optical regeneration in one- and two-pump parametric amplifiers using highly nonlinear optical fiber ,” IEEE Photon. Technol. Lett. 15 , 957 – 959 ( 2003 ). [CrossRef]  

6 . D. V. Kuksenkov , J. Rothman , M. Sauer , S. Li , and K. A. Hoover , ‘ 10 Gb/s 2R regeneration using fiber-optic limiting parametric amplifier ,’ European Conference on Optical Communication 2003, We 1.6.2 ( ECOC 2003, Rimini, Italy , 2003 ).

7 . K. K. Chow , C. Shu , Chinlon Lin , and A. Bjarklev , “ Polarization-insensitive widely tunable wavelength converter based on four-wave mixing in a dispersion-flattened nonlinear photonic Crystal fiber ,” IEEE Photon. Technol. Lett. 17 , 624 – 626 ( 2005 ). [CrossRef]  

8 . K. P. Hansen , J. R. Folkenberg , C. Peucheret , and A. Bjarklev , “ Fully dispersion controlled triangular-core nonlinear photonic crystal fiber ,” Optical Fiber Communication Conference 2003, PD2-1 ( OFC, Atlanta, Georgia , 2003 ).

9 . D. F. Grosz , C. Mazzali , S. Celaschi , A. Paradisi , and H. L. Fragnito , “ Modulation instability induced resonant four-wave mixing in WDM systems ,” IEEE Photon. Technol. Lett. 11 , 379 – 381 ( 1999 ). [CrossRef]  

10 . T. Hasegawa , K. Inoue , and K. Oda , “ Polarization independent frequency conversion by fiber four-wave mixing with a polarization diversity technique ,” IEEE Photon. Technol. Lett. 5 , 947 – 949 ( 1993 ). [CrossRef]  

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (6)

Fig. 1.
Fig. 1. Schematic illustration of pump-modulated four-wave mixing for extinct ratio enhancement. (a) The degraded input signal ω1 is in “one” state with ω2 as the CW source and ωc as the converted signal. (b) ω1 is in “zero” state with ωc suppressed.
Fig. 2.
Fig. 2. Experimental setup on pump-modulated four-wave mixing in a nonlinear photonic crystal fiber (PCF). The inset shows the microscopic picture of the mircrostructured region of the PCF. EDFA: erbium-doped fiber amplifier; PC: polarization controller; FFP: fiber Fabry-Perot filter.
Fig. 3.
Fig. 3. Optical spectra measured at different positions of the setup.
Fig. 4.
Fig. 4. Eye diagrams of (a) degraded 10 Gb/s RZ input signal and (b) the corresponding output.
Fig. 5.
Fig. 5. Plot of the bit-error-rate against the received optical power in a 10 Gb/s BER measurement.
Fig. 6.
Fig. 6. Plot of the output power and extinction ratio improvement against different output wavelength.

Equations (1)

Equations on this page are rendered with MathJax. Learn more.

P c ( L ) = ηγ 2 P 1 2 P 2 exp ( αL ) { [ 1 exp ( αL ) ] 2 α 2 }
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.