We show the first simultaneous OSNR monitoring of two 40 Gb/s OOK and DPSK channels, using only a wavelength selective switch and two slow photodetectors. Our approach is modulation format and bit-rate independent and can easily be included in existing reconfigurable networks.
© 2010 Optical Society of America
Current network infrastructure is moving away from traditional point-to-point networks to wavelength reconfigurable networks where different spectral channels can be reconfigurably rerouted to adapt to different bandwidth demands or to route around failure points in the network. Optical performance monitoring (OPM) techniques are of key importance for maintenance and diagnostics in these reconfigurable communication networks. In particular the ability to accurately and continuously monitor signal impairments at various points (nodes) of a network is highly desired [1,2] in order to locate and pinpoint impairment sources or failure locations in the network. The optical signal to noise ration (OSNR) is a key performance indicator of signal quality in communications networks.
Traditional techniques of determining the in-band OSNR via interpolation from the out-of band noise level fail as individual channels experience different impairments due to different paths through the network, and thus in-band noise can differ significantly from the noise level in-between channels. On the other hand, monitoring signal quality using digital signal processing techniques does provide accurate measurements. However it requires fast optic-to-electronic conversion and processing, which makes an implementation at every network node prohibitively expensive. It is therefore highly desired to develop novel techniques which do not rely on fast electronics and provide accurate measurements of the OSNR. Ideally such methods would be modulation format and symbol rate independent, so they can be easily adapted to different network requirements. Various impairment monitoring methods have been demonstrated based on both linear [3–5] and nonlinear optics [6–8].
One very promising scheme is based on interferometry , that takes advantage of the different coherence properties of signal and noise to determine the OSNR. This method is particularly advantageous due to its robustness against other impairments such as chromatic dispersion and polarization mode dispersion . A number of publications have shown OSNR measurements with this technique using various interferometer configurations ranging from free-space Michelson interferometers  to fiber-based and monolithic Mach-Zehnder interferometers [9, 10, 12, 13]. Furthermore it has also been demonstrated that an interferometer based approach can be used for simultaneous group velocity and polarization mode dispersion monitoring . However for integration into wavelength-routed networks, novel OSNR monitoring techniques should be easily reconfigurable, e.g. to adapt to different network plans, bit-rates and modulation-formats. They should be stable with respect to environmental conditions and should easily integrate with current network architecture, i.e. have low power and rack space requirements. All of the previous solutions require investment into new components, need to be stabilized against temperature and other environmental fluctuations and occupy additional rack space in the server rooms.
We demonstrate, for the first time, simultaneous multi-channel OSNR monitoring, by implementing the monitoring using a liquid crystal on Silicon (LCoS) based wavelength selective switch (WSS) and two slow photo-detectors. Because these devices are key instruments in reconfigurable add-drop multiplexers (ROADM) which are already present at the nodes of reconfigurable optical networks, the OSNR monitoring could be easily integrated with current network architecture. We take advantage of the high reconfigurability of the LCoS technology to create a delay line interferometer (DLI) inside the WSS. Thus we can measure the in-band OSNR by simply measuring the average output power at a designated output port. This OSNR monitor can be retrofitted to existing reconfigurable networks by a software update and attaching a slow photodiode to one of the output ports of the WSS. Therefore the additional rack space requirements are minimal and the system can be implemented with minimal infrastructure and maintenance investments. We demonstrate a measurement accuracy of better than 0.5 dB for OSNR values up to 30 dB for DPSK and up to 25 dB for NRZ-OOK.
The OSNR measurement technique is based on the different coherence properties of a modulated signal and amplified spontaneous emission (ASE) noise from optical amplifiers in the network, which is the main cause of OSNR degradation. A signal –even if modulated by a random bit pattern– remains partially coherent over several bit-periods. The coherence of ASE noise on the other hand is determined by the channel bandwidth and is in general significantly shorter than the signal coherence. Thus if the signal-only and noise-only coherence is known it is possible to calculate the OSNR of any combination . Here we implement a delay line interferometer to measure the coherence inside the WSS, by taking advantage of the advanced delay control capabilities of the WSS . This is achieved by splitting the LCoS such that we apply one delay to half of the light inside the WSS and a different delay to the other half. When the light interferes at the exit of the WSS an interferometer is created. The delay between the two arms of the interferometer can be between 0 and 40 ps and is limited by delay dependent loss. Similarly it is possible to vary the phase between the two arms, by varying the relative phase of the two phase gradients applied to the LCoS, thus changing between constructive and destructive interference. From a measurement of the ratio R of the power at constructive over the power at destructive interference it is possible to determine the OSNR via :
The experimental setup is depicted in Fig. 2. Two cw lasers (one external feedback (ECL), one distributed feedback (DFB) laser) centred at two adjacent wavelengths of the 100 GHz ITU grid are combined using a fibre coupler and data-encoded with either a NRZ-OOK or NRZ-DPSK pseudo-random bit sequence of length 231 – 1 using a Mach-Zehnder modulator. We add a specific amount of ASE noise from an Erbium-doped fibre amplifier, with its power being controlled by a variable optical attenuator.
The two channels pass through the WSS, where they are bandpass filtered to 100 GHz, and directed to two different output ports while we apply two different delays to the upper and lower half of the LCoS inside the WSS to create the DLI. The power after the DLI for each channel is measured with a slow photodiode at each of these ports. The reference OSNR is determined by measuring the power of signal only and noise only with the photodiode and no DLI. Typical average input powers of the signal-only for one channel into the WSS were between −10 and −15 dBm (dependent on modulation format and channel bandwidth).
Figure 3 shows the OSNR calculated from Eq. (1) as a function of reference OSNR for two (a) 40 Gbit/s NRZ-OOK signals and (b) 40 Gbit/s NRZ-DPSK signals measured simultaneously (circles and squares). We see a very good agreement between signal and reference OSNR with a measurement error below 0.5 dB up to 25 dB of OSNR in the case of OOK and up to 30 dB for DPSK. In both measurements the delay of the interferometer was 4 ps and the phase between the two arms was varied to determine the constructive and destructive average powers. The delay was chosen such that delay dependent loss of the interferometer was negligible and that the delay was sufficiently large so that the difference between n and s is sufficient (at very low delays (< 1 ps) the coherence function of the noise and signal are very similar which would degrade the monitor performance). A detailed analysis of the monitoring performance as a function of DLI delay is beyond the scope of this paper, however we have checked that values between ∼ 2 ps and ∼ 15 ps yield very similar results.
A large advantage of our OSNR monitoring implementation is that it can be readily reconfigured for a different channel bandwidth and a different DLI delay. Figure 3(c) shows a single channel measurement of a 40 Gbit/s NRZ-OOK signal at channel 23 (1551.32 nm) of the 50 GHz ITU grid. The slightly larger error can be attributed to the fact that the noise coherence is much closer to the signal coherence due to the reduced 50 GHz bandwidth, however the measurement error remains below 1 dB in the range from 5–25 dB OSNR.
A major advantage of this coherence-based OSNR monitoring scheme is its robustness against other impairments such as group velocity dispersion (GVD) or differential group delay (DGD) . This is illustrated in Fig. 4, which shows the measured OSNR after a 10 km single-mode fiber was inserted into the setup between the coupler and the WSS, after the signal-only coherence measurement was taken. This situation corresponds to the case when there is some residual dispersion after a reconfiguration of the network. The accuracy of the measurement is better than 0.5 dB over an OSNR range of 5–30 dB for both channels.
3. Discussion and conclusion
We have demonstrated simultaneous multichannel measurement of in-band OSNR of two 40 Gbit/s signals using only a WSS and a slow photodiode. Our approach is modulation format independent, robust to other impairments and can be readily reconfigured to different network layouts and channel bandwidths as demonstrated by the 50 GHz bandwidth measurement. The practical integration of the monitor into a network can be envisioned in two ways: Retrofitting the in-band OSNR measurements into existing WSSs via a software update and connecting a slow photodiode to one or more of the output ports. In this case the monitoring operation could be performed in a drop-and-continue fashion, where the LCoS is split three-ways so as to continue the majority of power along the transmission path and performing the DLI operation on a fraction of the power directed to an output port with a connected photodiode. As output ports on a WSS often are a scarce resource only a very low number of ports could be monitored simultaneously. Instead if monitoring of a large number of ports is desired it has to performed sequentially. Because the monitoring speed is limited by the update time of the WSS which is about 100 ms, monitoring a large number of channels possibly would take several seconds (two updates are required for each channel measurement). The second option is to use a separate WSS device as an OSNR monitor. Such a device would be located at a monitoring port, e.g. inside the ROADM. The disadvantage of this approach is, that a separate device is required. However it enables simultaneous monitoring of a relative large number of channels (the number of output ports of the WSS). Such a device could also directly incorporate the photodiodes.
In conclusion, we have presented a novel a method of integrating OSNR monitoring into a WSS device by implementing a DLI using the phase manipulation capabilities of the device. This approach offers significant advantages as it could either be integrated into existing networks, or as a standalone device for measuring OSNR of multiple channels simultaneously, thus significantly reducing investment, maintenance costs and rack space requirements.
The authors acknowledge the ARC linkage scheme with Finisar Australia for financial support and for providing the WSS.
References and links
1. Z. Pan, C. Yu, and A. E. Willner, “Optical performance monitoring for the next generation optical communication networks,” Opt. Fiber Technol. 16, 20–45 (2010). [CrossRef]
2. D. C. Kilper, R. Bach, D. J. Blumenthal, D. Einstein, T. Landolsi, L. Ostar, M. Preiss, and A. E. Willner, “Optical Performance Monitoring,” J. Lightwave Technol. 22, 294 (2004). [CrossRef]
3. B. Kozicki, A. Maruta, and K.-i. Kitayama, “Experimental demonstration of optical performance monitoring for RZ-DPSK signals using delay-tap sampling method.” Opt. Express 16, 3566–3576 (2008). [CrossRef] [PubMed]
4. J. Lee, D. Jung, C. Kim, and Y. Chung, “OSNR monitoring technique using polarization-nulling method,” IEEE Photon. Technol. Lett. 13, 88–90 (2001). [CrossRef]
5. 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, 177–183 (2007). [CrossRef]
6. T. T. Ng, J. L. Blows, M. Rochette, J. A. Bolger, I. Littler, and B. J. Eggleton, “In-band OSNR and chromatic dispersion monitoring using a fibre optical parametric amplifier,” Opt. Express 13, 5542 (2005). [CrossRef] [PubMed]
8. T. D. Vo, M. D. Pelusi, J. Schröder, F. Luan, S. J. Madden, D.-Y. Choi, D. A. P. Bulla, B. Luther-Davies, and B. J. Eggleton, “Simultaneous multi-impairment monitoring of 640 Gb/s signals using photonic chip based RF spectrum analyzer,” Opt. Express 18, 3938–3945 (2010). [CrossRef] [PubMed]
9. Z. Tao, Z. Chen, L. Fu, D. Wu, and A. Xu, “Monitoring of OSNR by using a Mach-Zehnder interferometer,” Microwave and Opt. Technol. Lett. 30, 63–65 (2001). [CrossRef]
10. X. Liu, Y.-H. Kao, S. Chandrasekhar, I. Kang, S. Cabot, and L. Buhl, “OSNR Monitoring Method for OOK and DPSK Based on Optical Delay Interferometer,” IEEE Photon. Technol. Lett. 19, 1172–1174 (2007). [CrossRef]
11. J. M. Oh, M. Brodsky, L. E. Nelson, G. Cadena, and M. D. Feuer, “Interferometric optical signal-to-noise ratio measurements of telecom signals with degraded extinction ratio,” Opt. Lett.33, 2065–2067 (2008). [CrossRef] [PubMed]
12. S.-W. Jeon, Y. H. Kim, B. H. Lee, M. A. Jung, and C.-S. Park, “OSNR monitoring technique based on cascaded long-period fiber grating with optically tunable phase shifter,” Opt. Express 16, 20603–20609 (2008). [CrossRef] [PubMed]
13. E. Flood, W. H. Guo, D. Reid, M. Lynch, A. L. Bradley, L. P. Barry, and J. F. Donegan, “In-band OSNR monitoring using a pair of Michelson fiber interferometers,” Opt. Express 18, 3618 (2010). [CrossRef] [PubMed]
14. Y. K. Lize, L. Christen, J.-Y. Yang, P. Saghari, S. Nuccio, A. E. Willner, and R. Kashyap, “Independent and Simultaneous Monitoring of Chromatic and Polarization-Mode Dispersion in OOK and DPSK Transmission,” IEEE Photon. Technol. Lett. 19, 3–5 (2007). [CrossRef]
15. M. A. F. Roelens, S. Frisken, J. A. Bolger, D. Abakoumov, G. Baxter, S. Poole, and B. J. Eggleton, “Dispersion Trimming in a Reconfigurable Wavelength Selective Switch,” J. Lightwave Technol. 26, 73–78 (2008). [CrossRef]