Power Transients in Hybrid Optical Amplifier (EDFA + DFRA) Cascades

Power Transients in Hybrid Optical Amplifier (EDFA + DFRA) Cascades Bárbara Dumas and Ricardo Olivares Electronic Engineering Department Universidad Técnica Federico Santa María Valparaíso, Chile bpilar.dumas@gmail.com, ricardo.olivares@usm.cl Abstract WRONs often involve channel drop/add procedures affecting the transmission quality due to the introduction of power excursions by the optical amplifiers. In this work, the Hybrid Amplifier (HA) transient response under WDM channel drop/add scenarios is analyzed by means of numerical simulation. When 17 out of 20 channels are dropped/added, the HA transient response is larger and faster for forward DFRA hybrid configurations. Two different reactions were identified: initial disturbances and steady state regions, both intensified by a cascade of 12 HAs. The amplifier s relative location determines power excursions up to the third amplification stage, being larger for DFRA + EDFA configurations. For the backward DFRA + EDFA configuration, 6.41 db to 8.45 db overshoots between the second and the sixth amplification stages were observed and the steady state power excursion reached 7.83 db. For the forward DFRA + EDFA configuration a 7.75 db overshoot and a 7.98 db steady state power excursion response in the fourth amplification stage were obtained. Keywords WRONs, EDFAs, DFRAs, HAs, transient response. I. INTRODUCTION Wavelength Division Multiplexed (WDM) networks use optical amplifiers to compensate both the optic fiber and the intermediate nodes losses. On one hand, Erbium Doped Fiber Amplifiers (EDFAs) are already a well established technology with advantageous properties such as high gain ( 30 db, [1]) and good power conversion efficiency (> 30%, [2]). On the other hand, Distributed Fiber Raman Amplifiers (DFRAs) are a recent and an attractive alternative due to their comparative advantages: more flexible and higher bandwidths (> 40 nm, [3]), lower equivalent noise figure ( 3 db, [4]), and they are less vulnerable to fiber nonlinearities than EDFAs. Considering hybrid amplification (EDFA + DFRA) on links that already use EDFAs is technically and economically attractive. Hybrid Amplifiers (HAs) combine both EDFA and DFRA advantages and mitigate their limitations, allowing the length and/or the transmission bandwidth of these links to be extended. In Wavelength Routed Networks (WRONs), optical amplifiers are subjected to a variable load (number of channels) due to the dynamic reconfiguration of links, the restoration by fiber cuts, or any other procedure involving the removal or the inclusion of WDM channels. EDFAs and DFRAs are vulnerable to this change in the number of transported channels, because of the Cross Gain Modulation (XGM) causing power transients that can significantly degrade the performance of these networks. Power transients and their control have been extensively studied in EDFAs [5] and in DFRAs [6]. Nevertheless, the transient response of HAs has been studied discreetly and in limited scenarios. For instance, in [7] the authors conclude that the transient response of HAs is dominated by the EDFA dynamic instead of DFRA s. For this reason, the proposed transient control method is implemented only for the EDFA, mitigating power excursions from 3 db to 0.6 db, when 10 out of 11 channels are dropped/added. However, there are no works referring to the impact of the amplifier s relative location on the HA transient response. Also, there are no reported works about the transient response along HA cascades accumulative effects. Hence, in this work, by means of numerical simulation, the transient response is analyzed in four configurations of HAs. In section II, the description of each hybrid configuration is presented. In section III, results are analyzed considering one amplification stage and cascades of 12 HAs. Finally, in section IV, the conclusions of this work are presented. II. HYBRID AMPLIFIER CONFIGURATIONS Four hybrid configurations are analyzed. Two of them locate the EDFA on first place followed by the DFRA ( EDFA + DFRA ). The other two configurations invert the order of the amplifiers ( DFRA + EDFA ). For both combinations DFRAs with backward and forward (f) pumping schemes are considered, as shown in Fig. 1(a) ( EDFA + DFRA b ), Fig. 1 ( EDFA + DFRA f ), Fig. 1(c) ( DFRA b + EDFA ) and Fig. 1(d) ( DFRA f + EDFA ). The propagation of 20 WDM channels evenly spaced (1 nm) from 1545 nm to 1565 nm is considered, with an input power of P s,in = 0 dbm/ch for every configuration. In each hybrid configuration the EDFA compensates, as a post-amplifier, 100 km of SMF (Standard Single-Mode Fiber) power losses which are estimated to be 20 db. Thus, the EDFA input power reaches P EDF A,in = 20 dbm/ch. The EDFA consists of a 12 m long EDF (Erbium Doped Fiber) pumped with 40 mw of power at 980 nm wavelength. Otherwise, the DFRA input P DF RA,in corresponds to the EDFA output for ( EDFA + DFRA ) configurations and P DF RA,in = P s,in for ( DFRA + EDFA ) configurations. The DFRA is designed to 453

The propagation equation of this model describes the power variation of signals and pumps regarding position and time. Both models are solved using numerical integration techniques. For the EDFA model the fourth order Runge-Kutta method is considered. For the DFRA model the finite difference method is applied and the relaxation method is used when considering the backward pump propagation. (a) III. NUMERICAL RESULTS A. Transient response of a single hybrid amplifier stage Fig. 2 shows the transient response for one of the three surviving channels, channel 11 (λ 11 = 1555 nm), for the hybrid configurations shown in Fig. 1(a) and Fig. 1, considering the drop/add of channels. 20 WDM channels are introduced at time t =0ms, 17 out of them (channels 1, 2, 3, 4, 5, 6, 8, 9, 10, 12, 13, 14, 16, 17, 18, 19 and 20) are removed at time t =1ms and then reinserted at time t =2ms. In order to compare the EDFA transient response with DFRA s, the curves associated to the DFRA transient response have been advanced 250 μs in time to correspond to the propagation delay of 50 km of Raman fiber. (c) (d) Figure 1. Hybrid amplifier configurations. compensate 50 km of SMF power losses, with two pumps located at 1455 nm and 1463 nm. When the backward pumping scheme is considered, pumps with power levels at 130 mw and 60 mw are used, alternatively, if the pumping scheme is reversed to the forward direction, pumps are set at 200 mw and 65 mw, respectively. The theoretical model used for the EDFA in the time domain is presented in [8]. This model neglects the ASE noise, because while working in high saturation levels this noise is negligible in relation with the signal power. The theoretical model used for the DFRA is presented in [9]. This model considers the effects of attenuation, Rayleigh scattering and stimulated Raman scattering (gain and depletion of channels and pumps). Anti-Stökes and Stökes spontaneous scatterings are not considered, since they are negligible with the channel spacing used in these simulations (1 nm) [6]. Figure 2. Transient response of EDFA + DFRA hybrid configurations. The EDFA transient response is amplified by the DFRA in both pumping schemes. This can be seen on larger steady state power excursions (from 3.11 db to 3.77 db and to 5.44 db, for backward and forward pumping schemes, respectively), on lower rising times (from 195 μs to192 μs and to 135 μs, for backward and forward pumping schemes, respectively) and hence on faster rising slopes (from 15.95 10 3 db/μs to 19.64 10 3 db/μs and to 40.30 10 3 db/μs, for backward and forward pumping schemes, respectively). These results can be explained though the XGM: immediately after the drop of the 17 channels the EDFA output power of the surviving channels grows due to the increased available power gain, which is now stimulated by the 3 surviving channels and distributed only among them (before the drop of channels the EDFA was stimulated by 20 channels and the 454

power gain was distributed among all of them). The steady state output power of channel 11 is determined by the current total input power (3 surviving channels) and by the EDFA saturation level. Therefore, the DFRA input power after the drop of channels corresponds to the surviving channels with the power excursions introduced by the EDFA. Thus, channel 11 is again amplified because of the increased available power gain, but to a lower intensity, because now the input power is higher and the DFRA is more saturated. Moreover, it can be seen that power excursions are lower when the DFRA pumping scheme is backward than when it is forward, because at the beginning of link the power gain is higher for the forward DFRA pumping scheme. Fig. 3 shows the power excursions of channel 11 at the output of the DFRA and the EDFA for the hybrid configurations shown in Fig. 1(c) and Fig. 1(d). EDFA corresponds to the EDFA output whose input has been previously amplified by a DFRA b and EDFA(f), by a DFRA f. Finally, after the 17 channels are re-added, the output power is re-established to the value before the drop of channels, for every hybrid configuration. That can be seen since steady state power excursions reach 0 db after t =2ms. With the aim to compare the transient response of every hybrid configuration, Fig. 4 shows the power excursions at the output of them all. It can be seen that the best and the worst response of the four hybrid configurations are EDFA + DFRA b and DFRA f + EDFA, respectively. This can be explained mainly for two reasons: first, the DFRA b restricts power excursions, because the energy transfer between pumps and signals is lower (than in a DFRA f ); and secondly, the relative position of the EDFA since it is more sensitive than the DFRA to changes in its total input power. Figure 4. Transient response of each hybrid configuration. Figure 3. Transient response of DFRA + EDFA hybrid configurations. Again, the previous amplifier transient response (DFRA) is amplified by the next amplifier (EDFA). For the EDFA, the steady state power excursion and the rising slope increase (from 1.09 db to 4.00 db and from 4.49 10 3 db/μs to14.34 10 3 db/μs, respectively), while the settling time is reduced (500 μs to 416 μs). In contrast, for the EDFA(f), the steady state power excursion and the settling time increase (from 3.86 db to 5.87 db and from 106 μs to456 μs, respectively), while the rising slope decreases (from 83.97 10 3 db/μs to 44.47 10 3 db/μs). It can be seen that power excursions are larger and faster for forward DFRA pumping scheme hybrid configurations compared to those that consider backward DFRA pumping scheme. On the other hand, both DFRAs (backward and forward) present faster and lower power excursions than the EDFA. It is also noted that the EDFA restricts the maximum steady state power excursion due its saturation level, lower than DFRA s. B. Hybrid amplifiers in cascade In order to observe the transient response in HA cascades, a 1800 km long link is considered, whose losses will be compensated by 12 HAs (150 km each), under the four presented hybrid configurations. The results (power excursion versus time) for channel 11, for the even amplification stages, are shown in Fig. 5 and Fig. 6, for configurations EDFA + DFRA and DFRA + EDFA, respectively, with backward (Fig. 5(a) and Fig. 6(a)) and forward (Fig. 5 and Fig. 6) DFRA pumping schemes. In Fig. 5(a) and Fig. 6(a) it is possible to identify two clearly distinct areas in the transient response of the different amplification stages. First, an initial disturbance (overshoot) can be seen, followed by a zone in which power excursions reach the new steady state. The latter corresponds to the power excursions reached by the surviving channels due to the power gain provided by the amplifier when stimulated only by them. As shown in these figures, power excursions of channel 11 increase for both transient and steady state areas, until the sixth and fourth amplification stages, respectively. From which power excursions decrease (overshoots and steady state values). 455

(a) (a) Figure 5. Transient response in cascades of 12 HAs, for configurations (a) EDFA + DFRA b and EDFA + DFRA f. Figure 6. Transient response in cascades of 12 HAs, for configurations (a) DFRA b + EDFA and DFRA f + EDFA. This behavior is due to long response times of both the EDFA and the DFRA b. Moreover, the input power from the second HA onwards corresponds to the power of the surviving channels with the transient characteristic accumulated along the previous amplification stages. Immediately after the drop of channels (in the transient area) each HA keeps providing the power gain stimulated by the 20 channels to the 3 surviving channels, those which, having a higher power than the achieved at steady state (because of the increased power gain introduced by the previous amplification stages), reach larger power excursions. The maximum power that these overshoots can achieve is related to the maximum power gain that the amplifier can provide under the stimulation of the 20 channels at steady state. Overshoot power excursions values go from 6.16 db and 6.41 db in the second amplification stage to 8.36 db and 8.45 db in the sixth amplification stage, for configura- tions EDFA + DFRA b and DFRA b + EDFA, respectively. For configurations EDFA + DFRA f and DFRA f + EDFA, overshoots reach 7.75 db, in the fourth amplification stage. Steady state power excursions (in the second identified area) increase along the cascade from 6.09 db and 7.74 db to 6.33 db and 7.83 db, between the third and the sixth amplification stages and then they remain constant, for configurations EDFA + DFRA b and DFRA b + EDFA, respectively. For configurations EDFA + DFRA f and DFRA f + EDFA, power excursions go from 7.09 db and 7.18 db to 7.71 db and 7.98 db, respectively, between the third and the sixth amplification stages, after which they remain constant. The steady state output power of each stage depends on the power gain provided by the amplifiers, which in turn depends on the input power of the surviving channels and on the amplifier saturation level. After the first HA, the input power of the surviving channels is being modified constantly by the transient response 456

introduced by the previous amplification stages. On the other hand, after the drop of channels, the total input power of each amplifier decreases and, with it, the amplifier saturation level is reduced. However, as the overshoots start to appear, the amplifier total input power increases, resulting a higher saturation level. Once these two parameters stabilize, power excursions of the surviving channels reach the steady state. As a way of comparison, Fig. 7 shows power excursions of channel 11 along cascades of 9 EDFAs, 18 DFRAs (for backward and forward pumping schemes) and 6 HAs (for the four hybrid configurations), covering a 900 km distance. Figure 7. Transient response in cascades of EDFAs, DFRAs and HAs. This figure shows that overshoots (undershoots) and rising slopes along the HA cascade reach intermediate values between those associated to EDFAs and DFRAs (for both pumping schemes) cascades. In addition, there is no difference between each backward or forward hybrid configuration, except for the two first amplification stages. IV. CONCLUSION A rigorous analysis of the transient response of HAs has been presented under an extreme scenario of drop/add of channels. DFRAs, in both backward and forward pumping schemes, have been considered, preceding and ahead of an EDFA. It could be seen that the amplifier s relative location and saturation level determine the intensity and speed of the power excursions of the surviving channels. Forward DFRA hybrid configurations presented larger and faster power excursions, especially when DFRA f is the first amplifier of the configuration. These transient effects are accumulated when HAs are cascaded. It was noted that cascades of amplifiers with longer response times, such as EDFAs, DFRAs and HAs (both last with backward DFRA pumping scheme), have shown transient responses where it is possible to identify an initial disturbance followed by a steady state condition. Immediately after the drop of channels, while the amplifiers are still providing the power gain stimulated by the 20 channels, the 3 surviving channels increase their power excursions generating overshoot disturbances. After the second amplifier, these channels will further increase their power excursions, because of the transient effects introduced by the previous amplification stages. In contrast, for cascades of amplifiers with shorter response time, such as forward DFRA hybrid configurations, the steady state is reached almost immediately. In addition, the results have shown that the EDFA increases the power excursions and the rising slopes of the transient zone. Meanwhile, the DFRA increases the steady state power excursions (since its saturation level is higher than EDFA s). Moreover, it was observed that power excursions are larger for the DFRA + EDFA hybrid configurations compared to those that invert the order of the amplifiers, both at the beginning of the transient as in the steady state region. However, the differences in the transient response between these configurations and EDFA + DFRA configurations are attenuated along the cascade. After the third amplification stage, steady state power excursions remain constant, and after the eighth amplification stage, initial disturbances start decrease. Finally, from the results shown in this work, the need to implement control techniques to mitigate or eliminate transient effects for the different configurations of HAs is clear. ACKNOWLEDGMENTS The authors would like to thank the partial support received from the UTFSM project DGIP-230964. REFERENCES [1] C. R. Giles and E. Desurvire, Modeling Erbium Doped Fiber Amplifiers, IEEE Journal of Lightwave Technology, vol. 9, pp. 271-283, 1991. [2] M. N. Islam, Raman Amplifiers for Telecommunications, IEEE Journal of Selected Topics in Quantum Electronics, vol. 8, pp. 548-559, 2002. [3] V. E. Perlin and H. Winful, On Distributed Raman Amplification for Ultrabroad-Band Long Haul WDM Systems, IEEE Journal of Lightwave Technology, vol. 20, pp. 409-416, 2002. [4] C. Rivera et al., Numerical Simulations and Experimental Results of a Hybrid EDFA-Raman Amplifier, International Microwave and Optoelectronics Conference (IMOC 2009), Brazil, 2009. [5] R. Olivares, A Comparative Study of Techniques to Control Power Transients in Optical WDM Networks, 10th European Conference on Networks & Optical Communications, England, 2005. [6] V. Pincheira et al., Per-band Link Control Transients Protection in Distributed Fiber Raman Amplifier Cascades, International Microwave and Optoelectronics Conference (IMOC 2009), Brazil, 2009. [7] J. H. Lee et al., A Detailed Experimental Study on Single-Pump Raman/EDFA Hybrid Amplifiers: Static, Dynamic and System Performance Comparison, IEEE Journal of Lightwave Technology, vol. 23, pp. 3484-3493, 2005. [8] S. R. Chinn, Simplified Modeling of Transients in Gain-Clamped Erbium Doped Fiber Amplifiers, Journal of Lightwave Technology, vol. 16, pp. 1095-1100, 1998. [9] M. Karásek and M. Menif, Channel Addition/Removal Response in All Optical Lumped Raman Amplifiers, IEEE Journal of Lightwave Technology, vol. 20, pp. 1680-1687, 2002. 457