We address the primary claim in the Comment by N. Alic et al. that our scheme for generating 1-μs tunable delays via Si-based waveguides in [Opt. Express 17, 7004-7010 (2009)] cannot support wavelength transparency by showing experimentally that the addition of a third conversion stage to reconvert to the input wavelength has minimal effect on the performance of our delay scheme.
© 2009 OSA
We are well aware of the importance of wavelength transparency for certain applications of all-optical delays as the first demonstration of wavelength preserving parametric optical delays was performed by our groups [2,3] and was subsequently followed by other experiments [4,5]. To realize wavelength preserving operation in our recently proposed scheme , an additional wavelength conversion stage is required. However as discussed in , the amount of tunable delay is completely independent from the third wavelength conversion. Both the tunable delay and dispersion management are fully enabled by tuning the wavelengths in fiber link I and II. Furthermore, wavelength preserving operation should not be considered as a universal requirement for tunable delay units. For example, tunable delays for the optical control of phased array antenna systems do not require fixed wavelengths. In addition, in future all-optical communication systems, wavelengths will be important resources for networking, and wavelength conversion should be viewed as a fundamental module. The output signal from the tunable delay stage should be converted to the desired wavelength (not necessarily to that of the input) by the wavelength conversion module following the delay stage. As a result several of the schemes that have been demonstrated [6,7] can find applications whether or not the reconversion stage is demonstrated.
The conversion efficiency in current silicon waveguides (−20 dB in our experiment) is not as high as that in highly nonlinear fibers, but nevertheless the induced power penalty is still acceptable. Indeed, a penalty below 0.5 dB has been demonstrated previously for wavelength conversion using Si waveguides at 10 Gb/s . We discuss this issue in  and predict that the third wavelength conversion, if needed for a particular application, will result in a similar power penalty. While such a prediction is based on well-known OSNR calculations, we show here that it is such the case via experimental verification with all three wavelength conversions in Si waveguides. The third wavelength conversion is performed with the original system for the 1-μs tunable delay as described in . The measured power penalties before and after the third conversion when λ 1 (the wavelength in fiber link I) is 1553.5 nm are plotted in Fig. 1 . Note that A PRBS with word length of 231-1 is used for the measurement.
Our data clearly show that the third wavelength conversion results in an additional power penalty to the system of ~0.5 dB, consistent with our discussion in . The three Si waveguides used in this experiment have on-off conversion efficiencies of approximately −20 dB, which is nearly identical to that of the two Si waveguides used in our original experiment. The output coupling from the Si waveguide to the single-mode fiber is slightly improved in this experiment, resulting in a 0.9-dB improvement in power penalty at a BER of 10−9 when compared to our original experiment. Our experiment demonstrates that, if wavelength preserving operation is required by the application, all three wavelength conversions can be accomplished using FWM in Si waveguides. Silicon devices have major advantages over highly nonlinear fibers in terms of conversion bandwidth, power consumption, photonic integration, and the absence of SBS. The main drawback of silicon currently is the low conversion efficiency. We note that silicon devices for parametric processes have been in development for approximately three years, whereas those based on fiber have been researched for nearly thirty years. The conversion efficiency of silicon waveguides is steadily improving and the reduction of coupling and propagation losses of the waveguide has lead to efficiencies of −10 dB , which will further decrease the power penalty of the delay system.
Lastly, in their Comment N. Alic et al., claim that our scheme “cannot be extended to considerably higher data rates”. This statement is counter to theoretical calculations and experimental facts. In , we show, for the first time, that not only can one achieve microsecond range continuous delays at 10 Gb/s experimentally, but also that the proposed scheme is theoretically capable of much higher data rates. Our analysis in  shows that in our scheme the signal distortion is determined by fourth-order fiber dispersion, whereas previous schemes were limited by third-order dispersion . Such an improvement allows for scalability to data rates up to 100 Gb/s at 1 µs delays. The scalability of the delay method to higher bit rates is also demonstrated by the recent experiment  which nearly duplicates our approach with the only differences being the third wavelength conversion for wavelength preserving operation and the use of highly nonlinear fibers.
References and links
1. Y. Dai, X. Chen, Y. Okawachi, A. C. Turner-Foster, M. A. Foster, M. Lipson, A. L. Gaeta, and C. Xu, “1 micros tunable delay using parametric mixing and optical phase conjugation in Si waveguides,” Opt. Express 17(9), 7004–7010 (2009). [CrossRef] [PubMed]
2. J. E. Sharping, Y. Okawachi, J. van Howe, C. Xu, Y. Wang, A. E. Willner, and A. L. Gaeta, “All-optical, wavelength and bandwidth preserving, pulse delay based on parametric wavelength conversion and dispersion,” Opt. Express 13(20), 7872–7877 (2005). [CrossRef] [PubMed]
3. A. Gaeta, J. E. Sharping, C. Xu, “Continuously tunable, pulse delay generator using wavelength conversion and dispersion,” United States Patent, 7538935.
4. E. Myslivets, N. Alic, J. R. Windmiller, R. M. Jopson, and S. Radic, “400 ns continuously tunable delay of 10 Gbps intensity modulated optical signal,” IEEE Photon. Technol. Lett. 21(4), 251–253 (2009). [CrossRef]
5. E. Myslivets, N. Alic, S. Moro, B. P. P. Kuo, R. M. Jopson, C. J. McKinstrie, M. Karlsson, and S. Radic, “1.56-micros continuously tunable parametric delay line for a 40-Gb/s signal,” Opt. Express 17(14), 11958–11964 (2009). [CrossRef] [PubMed]
6. Y. Wang, C. Yu, L. Yan, A. E. Willner, R. Roussev, C. Langrock, M. M. Fejer, J. E. Sharping, and A. L. Gaeta, “44-ns continuously tunable dispersionless optical delay element using a PPLN waveguide with two-pump configuration, DCF, and a dispersion compensator,” Photon. Technol. Lett. 19(11), 861–863 (2007). [CrossRef]
7. Y. Okawachi, M. A. Foster, X. Chen, A. C. Turner-Foster, R. Salem, M. Lipson, C. Xu, and A. L. Gaeta, “Large tunable delays using parametric mixing and phase conjugation in Si nanowaveguides,” Opt. Express 16(14), 10349–10357 (2008). [CrossRef] [PubMed]
8. B. G. Lee, A. Biberman, M. A. Foster, A. C. Turner, M. Lipson, A. L. Gaeta, and K. Bergman, “Bit-error-rate characterization of Silicon four-wave-mixing wavelength converters at 10 and 40 Gb/s,” CLEO 2008, paper CPDB4.
9. M. A. Foster, A. C. Turner, R. Salem, M. Lipson, and A. L. Gaeta, “Broad-band continuous-wave parametric wavelength conversion in silicon nanowaveguides,” Opt. Express 15(20), 12949–12958 (2007). [CrossRef] [PubMed]