An in-band optical signal-to-noise ratio (OSNR) monitor is proposed, based on an instantaneous polarization state distribution analysis. The proposed monitor is simple, and is applicable to polarization division multiplexed signals. We fabricate a high-speed Stokes polarimeter that integrates a planar lightwave circuit (PLC) based polarization filter, high-speed InP/InGaAs photodiodes and InP hetero-junction bipolar transistor (HBT) trans-impedance amplifiers (TIA). We carry out proof-of-concept experiments with the fabricated polarimeter, and successfully measure the OSNR dependent polarization distribution with 100-Gb/s dual polarization quadrature phase shift keying (DP-QPSK) signals.
© 2012 OSA
The in-band optical signal-to-noise ratio (OSNR), which provides a quantitative measure of the amplified spontaneous emission (ASE) level, is one the key parameters when we design and maintain optical transport systems. Until now, the in-band OSNR has been measured with a linear interpolation method , which interpolates the noise level at a channel wavelength from the out-of-band noise level. However, it is difficult to apply the interpolation method to optical signals in reconfigurable optical add/drop multiplexer (ROADM) based networks, because the noise is shaped by optical filters. It is also difficult to apply the method to dense wavelength-division-multiplexed high-speed signals whose baud rate is comparable to the channel spacing, such as a 100-Gbps dual polarization quadrature phase shift keying (DP-QPSK) signal with a 50-GHz channel spacing, because the signal spectrum overlaps the noise floor .
Several methods have been proposed for monitoring in-band OSNR, including polarization nulling , interferometry measurement of the amplitude autocorrelation function , orthogonal heterodyne mixing  and coherent detection with digital signal processing . Of these, polarization nulling based on polarization analysis is simple, but it is considered incompatible with a polarization division multiplexed (PDM) signals, such as a DP-QPSK signal. The coherent detection is a powerful tool, but it requires a local oscillator and accompanying digital signal processing, including a frequency offset compensation, timing synchronization, FFT and two polarization recovery.
In this paper, we propose a new in-band OSNR monitoring method based on high-speed polarization analysis. It utilizes the difference between the instantaneous polarization states of signals and noise. The proposed method is based on an asynchronous polarization measurement without a local oscillator. To confirm its operating principle, we fabricated a high-speed integrated-type Stokes polarimeter, and carried out proof-of-concept experiments with 100-Gb/s DP-QPSK signals.
2. Principle of OSNR monitor
The polarization nulling method is based on the assumption that a signal is fully polarized while ASE noise is fully unpolarized. A PDM signal such as a DP-QPSK signal contains two orthogonal polarization states with equal amplitudes, and its time-averaged degree of polarization (DOP) is zero as shown in Fig. 1(a) . However, if we use a high-speed Stokes polarimeter that is faster than the signal speed, we can observe its square-shape and lens-like instantaneous polarization distribution as shown in Fig. 1(b) .
Similarly, although the time-averaged DOP of ASE noise is zero as shown in Fig. 2(a) , if we pass the ASE noise through a narrow-band optical filter, and observe its polarization state with a high-speed polarimeter whose bandwidth is similar to or wider than that of the optical filter, we can observe its instantaneous polarization state distribution as shown in Fig. 2(b). Figures 1(b) and 2(b) indicate that we can distinguish the signal from the noise, and estimate the OSNR from the polarization state distribution.
As a measure of the OSNR, we propose using the standard deviation σ of the “thickness” of the polarization state distribution. In the rest of this section, we derive the relationship between OSNR and σ.
We denote the Jones vector fed into the polarimeter asEqs. (5), (6) and (7), we finally obtain an expression of the standard deviation asEquation (7) indicates that we can calculate the OSNR from the σ value of the thickness of the polarization state distribution. It should be noted that chromatic dispersion (CD) and polarization mode dispersion (PMD) affect the OSNR estimation accuracy via σ, as they deform the polarization state distribution. Further work will be needed to make the estimated OSNR insensitive to CD and PMD.
3. Design and fabrication of high-speed polarimeter
We designed and fabricated a high-speed Stokes polarimeter, which is a key component of our proposed OSNR monitor. Figure 3(a) shows the schematic configuration of our fabricated polarimeter, which consists of a silica-based planar lightwave circuit (PLC) polarization filter , InP/InGaAs photodiode (PD) arrays and InP HBT trans-impedance amplifiers (TIA). These building blocks are integrated in a surface mount technology (SMT) package as shown in Fig. 3(b). The basic configuration is the same as that of a coherent receiver , except for the PLC polarization filter.
The input signal is divided into two orthogonal polarization components at a polarization beam splitter (PBS) with quarter waveplates . After the PBS, the polarization states are aligned by a polarization rotator. Both components are then tapped by 2:1 couplers, and the tapped components are fed to TIA1 via PDs. The rest of the components are mixed in an optical 90-degree hybrid and fed to TIA2 and TIA3 via PDs. When we denote the outputs from TIA1, TIA2 and TIA3 as I1, I2 and I3, respectively, they are proportional to Stokes parameters as follows
The polarization filter was fabricated using a PLC with a refractive index difference of 1.5%. The chip size was 13.5 x 19 mm. The polarization extinction ratio of the PBS was better than 20 dB over the C and L bands. The 3-dB bandwidth of the fabricated polarimeter was around 19 GHz for S1, S2 and S3, as shown in Fig. 4 . Figure 5 shows the polarization state of a 32 GBaud DP-QPSK signal measured with the fabricated polarimeter. We confirmed that the fabricated polarimeter could measure the polarization state change of a 32 GBaud signal.
We measured the polarization state distribution of a 100-Gb/s DP-QPSK signal for different OSNRs. The experimental setup is shown in Fig. 6(a) . The outputs from the polarimeter were measured with a sampling oscilloscope at 50-Gsample/s. Here it should be noted that the optical filter bandwidth is 1 nm, and is wider than that of the polarimeter. However, it was possible to measure the partial polarization state distribution of ASE noise. Figures 6(b), 6(c) and 6(d) show the polarization state distributions observed at OSNRs of 10, 14 and 18 dB, respectively. We found that the obtained polarization state distribution depends strongly on the OSNR.
We then measured the standard deviation σ of the thickness of the polarization state distribution while changing the OSNR for various input polarization states. Figure 7(a) shows the OSNR dependent σ. The theoretical curve obtained from Eq. (7) is also shown in the figure, where we used γ = 0.0925, ε = 0.52 and σ0 = 0.063. We confirmed that the measured σ agreed with the theoretical curve. The OSNR estimated from Fig. 7(a) is shown in Fig. 7(b) as a function of the given OSNR. The measurement accuracy is within 2-dB up to an OSNR of 20 dB. We believe the polarization dependent accuracy seen in Fig. 7(b) could be improved by precise calibration and the use of narrow-band filter.
We proposed an OSNR monitoring method based on high-speed polarization analysis. We fabricated a high-speed integrated Stokes polarimeter consisting of a silica-based PLC polarization filter, InP/InGaAs PDs and InP HBT TIAs, and successfully confirmed the operating principle of the proposed method with the fabricated polarimeter.
We thank Y. Miyamoto, T. Goh, Y. Nasu, S. Kodama and H. Takahashi for helpful discussions and encouragement.
References and links
1. H. Suzuki and N. Takachio, “Optical quality monitor built into WDM linear repeaters using semiconductor arrayed waveguide grating filter monolithically integrated with eight photodiodes,” Electron. Lett. 35(10), 836–837 (1999). [CrossRef]
2. E. Flood, W. H. Guo, D. Reid, M. Lynch, A. L. Bradley, L. P. Barry, and J. F. Donegan, “Interferometer based in-band OSNR monitoring of single and dual polarisation QPSK signals,” in Proc. ECOC’11, Th.9.C.6 (2010).
3. J. H. Lee, H. Y. Choi, S. K. Shin, and Y. C. Chung “A review of the polarization-nulling technique for monitoring optical-signal-to-noise ratio in dynamic WDM network,” J. Lightwave Technol. 24(11), 4162–4171 (2006). [CrossRef]
4. C. Xie, D. C. Kilper, L. Möller, and R. Ryf, “Orthogonal-polarization heterodyne OSNR monitoring insensitive to polarization-mode dispersion and nonlinear polarization scattering,” J. Lightwave Technol. 25(1), 177–183 (2007). [CrossRef]
5. T. B. Anderson, S. Chen, A. Tran, D. F. Hewitt, L. B. Du, and A. J. Lowery, “Optical performance monitoring with low bandwidth coherent receivers,” in Proc. ECOC’12, Th.2.A.1 (2012).
6. B. Szafraniec, “Performance monitoring and measurement techniques for coherent systems,” in Proc. OFC’12, OW4G.5 (2012).
7. T. Saida, Y. Orihara, H. Yamada, K. Takiguchi, T. Goh, and K. Okamoto, “Integrated optical polarization analyser on planar lightwave circuit,” Electron. Lett. 35(22), 1948–1949 (1999). [CrossRef]
8. K. Murata, T. Saida, K. Sano, I. Ogawa, H. Fukuyama, R. Kasahara, Y. Muramoto, H. Nosaka, S. Tsunashima, T. Mizuno, H. Tanobe, K. Hattori, T. Yoshimatsu, H. Kawakami, and E. Yoshida, “100-Gbit/s PDM-QPSK coherent receiver with wide dynamic range and excellent common-mode rejection ratio,” Opt. Express 19(26), B125–B130 (2011). [CrossRef] [PubMed]
9. Y. Nasu, T. Mizuno, R. Kasahara, and T. Saida, “Temperature insensitive and ultra wideband silica-based dual polarization optical hybrid for coherent receiver with highly symmetrical interferometer design,” Opt. Express 19(26), B112–B118 (2011). [CrossRef] [PubMed]