Phase-sensitive amplifiers (PSA) using periodically poled (PPLN) LiNbO3 waveguides are promising as low-noise optical amplifiers. However, it is difficult to realize in-line operation for multi-level phase modulated signals using a PPLN based PSA with the conventional configuration. In this paper, we report a PPLN based in-line PSA that can regenerate quadrature phase shift keying (QPSK) signals. Multi-stage frequency mixing in a multiple quasi-phase matched LiNbO3 waveguide allows carrier phase recovery from a QPSK signal. Non-degenerate parametric amplification enables the phase-sensitive amplification of a QPSK signal. Amplitude and phase regeneration is examined utilizing gain saturation and phase squeezing capability.
© 2014 Optical Society of America
A high signal-to-noise ratio (SNR) is the most important factor in terms of achieving high capacity optical communication with high spectral efficiency. Recently, it has been pointed out that the SNR or signal quality is limited by nonlinear noise, which is caused by the optical Kerr effect in transmission fiber and amplified spontaneous emission (ASE) from optical amplifiers . The phase-sensitive amplifier (PSA) is attracting a lot of interest because it has the potential to mitigate both intensity noise and phase noise. There have been many recent studies on χ(3) based PSA using highly nonlinear fiber [2–6]. Meanwhile, recent advances on periodically poled LiNbO3 (PPLN) waveguides enable us to explore a χ(2) based counterpart [7–10]. The potential advantages of a PPLN based PSA are its compactness due to its high nonlinearity, its negligible chirping and negligible stimulated Brillouin scattering both due to the low χ(3) and its potential for the integration of several functions on a chip. High capacity communications will require the use of spectrally efficient multilevel phase modulation such as quadrature phase shift keying (QPSK) and quadrature amplitude modulation (QAM). There are several problems that must be overcome if we are to realize PSA for such multilevel phase modulation. One is that a conventional degenerate PSA amplifies an in-phase signal and de-amplifies a quadrature signal thus making it unsuitable for multilevel phase modulation. In this paper, the term degenerate refers to the signal and idler frequency. A solution to this problem has been demonstrated by utilizing non-degenerate parametric amplification in which a phase conjugated idler is also input into the PSA . Recently, the mitigation of nonlinear phase noise using a non-degenerate fiber-based PSA has been demonstrated for both QPSK and 16 QAM [11,12]. The amplification of 16 QAM using a PPLN-based PSA has also been demonstrated using a non- degenerate configuration . The second problem is that we must extract the carrier phase from the incoming multilevel phase modulated signal and generate a pump that is locked to the carrier phase. A solution for the second problem been demonstrated with a fiber based PSA, which utilizes multi-stage frequency mixing for carrier phase recovery from a QPSK signal . Another problem is that if we are to construct an in-line amplifier with only a signal input, we must generate a phase conjugated idler from the input signal. A multi-stage frequency mixing approach also enables the generation of a phase conjugated signal . However, an additional idea is required if we are to incorporate the same principle of multi-stage frequency mixing in a PPLN based PSA, because the precise phase matching characteristics of a PPLN with a single poling pitch do not allow multi-stage frequency mixing.
Recently, we demonstrated a PPLN based in-line PSA for QPSK signals experimentally . In this paper, we discuss the operation of this PSA in detail. We fabricated a multiple-quasi-phase matched (M-QPM) LiNbO3 waveguide that enables multi-stage frequency mixing in a single chip . Multi-stage frequency mixing in a multiple quasi-phase matched LiNbO3 enables carrier phase recovery from the QPSK signal. Non-degenerate parametric amplification enables the phase-sensitive amplification of QPSK signals. We examined amplitude and phase regeneration utilizing gain saturation and the phase squeezing capability.
2. Configuration of in-line PSA for QPSK signal
Figure 1 shows the configuration of an in-line PSA for QPSK signals. We use three PPLN waveguides, one for carrier recovery, one for pump generation, and one for parametric amplification. All the PPLN waveguides are fabricated by using direct bonding and dry etching to obtain high efficiency . All the PPLN waveguides are assembled in a fiber pigtail module, which enables us to input or output the 1.55 μm band signal and 0.78 μm band pump separately by utilizing a built-in dichromatic mirror . The QPSK signal and local oscillator1 (LO1) are combined with a fiber coupler, amplified with an EDFA, and input into PPLN1, which is an M-QPM device. The signal wavelength is 1535.04 nm and the LO1 wavelength is 1534.25 nm. Figure 2(a) shows the phase matching curve of the M-QPM device. In this device we modulated the spatial phase of a periodically poled structure. The phase modulation function was optimized to obtain three QPM peaks with a 100 GHz spacing .
The M-QPM device enables us to utilize multi-stage frequency mixing as illustrated in Fig. 2(b). The signal is converted to 0.78 μm band light with a second harmonic generation (SHG) process, and idler 1 is generated through difference frequency generation (DFG) between the SH light and LO1. The wavelength of idler 1 matched the second peak of PPLN1 and so idlers 2 and 3 are subsequently generated by SHG/DFG processes. The phase of each idler can be written as follows.4]. We examined the carrier phase recovery using a 20 Gbaud QPSK signal. Figure 2(c) shows the output spectra of PPLN1. Idler 3 exhibits a line spectrum as a consequence of the carrier phase recovery. Idler 3 was extracted using a liquid crystal on silicon (LCOS) filter and injected into a laser diode to obtain phase locked local oscillator 2 (LO2) as shown in Fig. 1. LO1 and LO2 are combined with a fiber coupler, amplified with an EDFA and injected into PPLN2 whose SHG phase matching wavelength is matched with idler 1. A 0.78 μm band pump is generated through the sum frequency generation (SFG) of LO1 and LO2 in PPLN2. The 0.78 μm band pump is injected into PPLN3 whose SHG phase matching wavelength is identical to that of PPLN2. If the phase matching condition is chosen for the SHG of frequency ω to 2ω, the phase matching conditions for parametric interaction between 2ω, ω−Δω, and ω + Δω are maintained over 30 nm of Δω detuning considering the dispersion of LiNbO3. The signal and idler 2 outputs from PPLN1 are extracted using an LCOS filter and injected into PPLN3 for non-degenerate optical parametric amplification (OPA). The phase of the input is multiplied three times in idler 2, which results in a complex conjugate of the QPSK signal (i.e. 0, π/2, π, 3π/2 are converted to 0, -π/2, -π, −3π/2, respectively).
The phase of the 0.78 μm band pump is given as follows.
3. Regeneration of QPSK signal
The phase-sensitive performance can be understood in detail by using the following analysis. The electric field of the signal output from PPLN3 is given by the following equation [5,17]Figure 3(a) and 3(b), respectively,show the calculated gains for the signal and output phases of the PSA assuming gL = 1.3, and signal to idler 2 intensity ratios |EI2|2/|Es|2 of 0.2, 0.4, and 0.6. The maximum gain can be obtained at a multiple of the π/2 phase variation as shown in Fig. 3(a). Note that as the idler power is increased, a higher gain and a higher phase-sensitive dynamic range can be obtained. The output phase is regenerated to give a π/2 step variation as shown in Fig. 3(b) by utilizing the non-degenerate parametric amplification of the signal and idler 2. As shown in Fig. 3, even though a higher idler intensity leads to a higher phase sensitive gain and deeper de-amplification, the flatness of the output phase worsens. So there is an optimum intensity ratio value between signal and idler 2 that minimizes the phase variation of the output. Such optimization was discussed for the phase regeneration of multilevel phase coded signals using non-degenerate FWM . According to the reference, the optimum signal to idler intensity ratio |EI2|2/|Es|2 for the regeneration of QPSK signals is 0.25.
Figure 4 shows the output spectra of PPLN3 with and without 0.78 μm band pump injection. In our experiment, signal gain fluctuation was caused by the phase drift due to the acoustic vibration of the fiber optics. The gain fluctuation is evidence for the phase-sensitive nature of the non-degenerate parametric interaction. We used a phase locked loop (PLL) based on a LiNbO3 phase modulator and a piezoelectric fiber stretcher to compensate for the phase drift and to lock the relative phase to give maximum or minimum gain . We obtained a signal gain of 10 dB and a phase-sensitive dynamic range exceeding 6 dB in this case. The signal to idler intensity ratio |EI2|2/|Es|2 was 0.35 in our experiment. This means that there is a possibility of improving the phase regeneration performance by using a higher signal power if we reduce the signal attenuation in the LCOS device. To balance the intensity of signal and idler, an additional attenuation of 22 dB was delivered by the LCOS device. It is difficult to predict the parametric gain coefficient gL from the experimental result, because the signal gain also depends on the idler intensity. From a comparison of the measured gain of 10 dB and Fig. 3(a), we estimate that gL was slightly larger than 1.3.
We attempted QPSK signal regeneration using the phase-sensitive gain. Figure 5 shows the experimental setup. A 20 Gbaud QPSK signal was generated by using an external cavity laser diode (ECLD) and a QPSK LiNbO3 modulator driven by two 20 Gbit/s pseudo random bit sequences generated by a pulse pattern generator (PPG). For the phase regeneration experiment, artificial phase noise was added to the QPSK signal by using a phase modulator driven by a sinusoidal wave at a frequency of 4.2 GHz. For the amplitude regeneration experiment, artificial amplitude noise instead of phase noise was added to the QPSK signal by using an intensity modulator driven by a sinusoidal wave at a frequency of 4.2 GHz. The QPSK signal with added noise was amplified with an in-line PSA and analyzed with a digital coherent receiver.
Figure 6 shows constellation diagrams observed with a digital coherent receiver with several levels of superimposed phase noise. As seen in Fig. 6(a), 6(c) and 6(e), the phase variance is larger than the amplitude variance in the input signals. In contrast, the phase variance and amplitude variance are almost the same in the output signals as shown in Fig. 3(b), 3(d), and 3(f). These results confirm the phase squeezing capability of the PSA. Although the absolute phase spread of the input and output is almost the same, this might be attributable to the non-optimized idler/signal intensity ratio, signal attenuation, and additional noise generated in the EDFA used in the idler generation. We also observed a slight degradation in the amplitude noise. This is due to the conversion loss of idler 2, signal attenuation, and ASE noise added by the EDFA. Further improvement of the idler generation efficiency should remedy the problem of amplitude noise degradation.
Another approach for reducing the amplitude noise is to utilize the gain saturation of the PSA [8,18]. We have already demonstrated that the gain saturation of the PSA is useful for suppressing amplitude noise in a recirculation loop transmission experiment for a BPSK signal . In this experiment, the signal and idler 2 are boosted with an additional EDFA and input into an OPA module to obtain gain saturation as illustrated in Fig. 1. Figure 7 shows the gain as a function of the total power input into the PPLN3 module. As the pump power is transferred to the output power of the signal and idler, the parametric gain becomes saturated. The gain responds almost instantaneously to the variation in the input intensity, and this intensity variation can be reduced by utilizing the gain saturation. We have obtained a gain saturation of up to 6 dB by increasing the input power, as shown in Fig. 7.
Artificial intensity noise is added to the signal as shown in Fig. 5. Figure 8 shows input and output constellation diagrams with different input powers. As can be seen in Fig. 8(a), the amplitude variance is larger than the phase variance in the input signal. By contrast, although we observe additional amplitude noise resulting from the signal attenuation caused by the LCOS device and insertion of the EDFA, the phase variance and amplitude variance are almost the same in the output signals as shown in Fig. 8(b) and (c). In addition, as the input power is increased, less amplitude noise is observed. Although the SNR must be improved to achieve a reduction in absolute amplitude noise, the gain saturation appears to be effective in mitigating the amplitude noise. p. These results suggest that gain saturation in the PSA is also useful for mitigating the amplitude noise of a QPSK signal in a long haul transmission.
We demonstrated a PPLN-based in-line PSA for QPSK signals. Carrier phase recovery using an M-QPM device and non-degenerate parametric amplification enabled us to realize an in-line PSA for QPSK. We examined amplitude and phase regeneration utilizing gain saturation and a phase squeezing capability. Although we demonstrated the capacity for reducing phase and amplitude noise, we can expect further improvement by reducing the conversion loss of the idler within the PSA.
This work was supported in part by JSPS KAKENHI Grant Number 25889052.
References and links
1. A. Bononi, P. Serena, and N. Rossi, “Nonlinear signal-noise interactions in dispersion-managed links with various modulation formats,” Opt. Fiber Technol. 16(2), 73–85 (2010). [CrossRef]
2. R. Slavík, F. Parmigiani, J. Kakande, C. Lundström, M. Sjödin, P. A. 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]
3. Z. Tong, C. Lundström, P. A. Andrekson, C. J. McKinstrie, M. Karlsson, D. J. Blessing, E. Tipsuwannakul, B. J. Puttnam, H. Toda, and L. Grüner-Nielsen, “Towards ultrasensitive optical links enabled by low-noise phase-sensitive amplifiers,” Nat. Photonics 5(7), 430–436 (2011). [CrossRef]
4. J. Kakande, A. Bogris, R. Slavik, 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 Proceedings of the 36th European Conference on Optical Communications (ECOC 2010) PD 3.3 (2010).
5. J. Kakande, R. Slavík, F. Parmigiani, A. Bogris, D. Syvridis, L. Grüner-Nielsen, R. Phelan, P. Petropoulos, and D. J. Richardson, “Multilevel quantization of optical phase in a novel coherent parametric mixer architecture,” Nat. Photonics 5(12), 748–752 (2011). [CrossRef]
6. K. Croussore and G. Li, “Phase regeneration of NRZ-DPSK signals based on symmetric-pump phase-sensitive amplification,” IEEE Photon. Technol. Lett. 19(11), 864–866 (2007). [CrossRef]
7. K. J. Lee, F. Parmigiani, S. Liu, J. Kakande, P. Petropoulos, K. Gallo, and D. Richardson, “Phase sensitive amplification based on quadratic cascading in a periodically poled lithium niobate waveguide,” Opt. Express 17(22), 20393–20400 (2009). [CrossRef] [PubMed]
8. Á. Szabó, B. J. Puttnam, D. Mazroa, S. Shinada, and N. Wada, “Investigation of an all-optical black-box PPLN-PPLN BPSK phase regenerator,” IEEE Photon. Technol. Lett. 24(22), 2087–2089 (2012). [CrossRef]
9. M. Asobe, T. Umeki, and O. Tadanaga, “Phase sensitive amplification with noise figure below the 3 dB quantum limit using CW pumped PPLN waveguide,” Opt. Express 20(12), 13164–13172 (2012). [CrossRef] [PubMed]
11. B. Corcoran, S. L. I. Olsson, C. Lundström, M. Karlsson, and P. A. Andrekson, “Mitigation of nonlinear impairments on QPSK data in phase-sensitive amplified links,” in Proceedings of the 39th European Conference on Optical Communications (ECOC 2013) We.3.A.1 (2013). [CrossRef]
12. S. L. I. Olsson, T. A. Eriksson, C. Lundström, M. Karlsson, and P. A. Andrekson, “Linear and nonlinear transmission of 16-QAM over 105 km phase-sensitive amplified link,” in Proceedings of the Optical Fiber Communication Conference (OFC) 2014 Th1H.3 (2014). [CrossRef]
13. T. Umeki, O. Tadanaga, M. Asobe, Y. Miyamoto, and H. Takenouchi, “First demonstration of high-order QAM signal amplification in PPLN-based phase sensitive amplifier,” in Proceedings of the 39th European Conference on Optical Communications (ECOC 2013) PD1.C.5, (2013). [CrossRef]
14. M. Asobe, T. Umeki, H. Takenouchi, and Y. Miyamoto, “In-line phase-sensitive amplifier for QPSK signal using multiple QPM LiNbO3 waveguide,” in Proceedings of the 18th OptoElectronics and Communications Conference/Photonics in Switching 2013 (OECC/PS 2013) PD 3.3 (2013).
15. M. Asobe, O. Tadanaga, H. Miyazawa, Y. Nishida, and H. Suzuki, “Multiple quasi-phase-matched device using continuous phase modulation of χ(2) grating and its application to variable wavelength conversion,” IEEE J. Quantum Electron. 41(12), 1540–1547 (2005). [CrossRef]
16. T. Umeki, O. Tadanaga, and M. Asobe, “Highly efficient wavelength converter using direct-bonded PPZnLN ridge waveguide,” IEEE J. Quantum Electron. 46(8), 1206–1213 (2010). [CrossRef]
17. J. Kakande, R. R. Slavík, F. Parmigiani, P. Petropoulos, and D. J. Richardson, “All-optical processing of multi-level phase shift keyed signals, in Proceedings of the Optical Fiber Communication Conference /National Fiber Optic Engineers Conference (OFC/NFOEC) 2012 OW1I.3 (2012).
18. C. Lundström, B. Corcoran, M. Karlsson, and P. A. Andrekson, “Phase and amplitude characteristics of a phase-sensitive amplifier operating in gain saturation,” Opt. Express 20(19), 21400–21412 (2012). [CrossRef] [PubMed]
19. T. Umeki, M. Asobe, H. Takara, Y. Miyamoto, and H. Takenouchi, “Multi-span transmission using phase and amplitude regeneration in PPLN-based PSA,” Opt. Express 21(15), 18170–18177 (2013). [CrossRef] [PubMed]