Expand this Topic clickable element to expand a topic
Skip to content
Optica Publishing Group

Single-PPLN-assisted wavelength-/time-selective switching/dropping/swapping for 100-GHz-spaced WDM signals

Open Access Open Access

Abstract

We propose an approach to implementing wavelength- and time-selective optical switching, dropping and swapping based on the sum-frequency generation (SFG) or cascaded sum- and difference-frequency generation (cSFG/DFG) in a periodically poled lithium niobate (PPLN) waveguide. Analytical solutions are derived, showing the parametric depletion effect for optical switching and the narrow-band operation due to quasi-phase matching (QPM) condition of PPLN. Using parametric depletion effect of SFG process, we demonstrate wavelength- and time-selective optical switching for ITU-grid compatible 40-Gbit/s wavelength-division multiplexed (WDM) signals with a channel spacing of 100 GHz. Less than 1-dB power penalty at a bit-error rate (BER) of 10−9 is measured for the wavelength- and time-selective switching channel. Negligible impacts are observed on other channels of WDM signals. Using combined effects of parametric depletion and wavelength conversion of cSFG/DFG processes, we demonstrate wavelength- and time-selective optical dropping for ITU-grid compatible 100-GHz-spaced 40-Gbit/s WDM signals. Moreover, we demonstrate optical swapping between two 100-GHz-spaced 40-Gbit/s signals. The obtained theoretical and experimental results confirm single-PPLN-assisted wavelength- and time-selective optical switching, dropping and swapping for 100-GHz-spaced WDM signals, which might potentially be extended to WDM signals with narrower channel spacing.

©2013 Optical Society of America

1. Introduction

Optical signal processing with ultrafast response is a promising technique to enable fast signal manipulation in the optical domain without optical-electrical-optical (OEO) conversion. Rapid increase of traffic rates has driven the demand for different optical signal processing functions at network nodes to reduce the latency, cost and power consumption. Optical nonlinearities on different platforms, e.g., highly nonlinear fiber (HNLF), semiconductor optical amplifier (SOA), periodically poled lithium niobate (PPLN) waveguide, chalcogenide (As2S3) waveguide, and silicon waveguide, are potentially suitable candidates to facilitate miscellaneous optical signal processing operations [19].

Among lots of optical signal processing functions, optical switching, dropping and swapping are promising ones required for future high-speed optical communication systems. When referred to the switching in the optical domain, optical switching employs one light to control another data-carrying light “ON” or “OFF”. Optical dropping removes one data-carrying channel and passes it to another channel. Optical swapping refers to the direct information exchange between two data-carrying channels. Previously, optical switching, dropping and swapping were widely reported using various optical nonlinearities in fibers or integrated waveguides, including the use of four-wave mixing (FWM) in HNLFs [1014], two-photon absorption (TPA) in microring resonators [15, 16], second-order nonlinearities and their cascading in PPLN waveguides [1722], cross-absorption modulation (XAM), cross-phase modulation (XPM) or Raman effect in silicon waveguides [2325]. Those prior art works have shown impressive operation performance of switching/dropping/swapping. However, to the best of our knowledge, there has been little research efforts on wavelength- and time-selective optical switching, dropping and swapping of wavelength-division multiplexed (WDM) signals with a narrow channel spacing such as 100-GHz (~0.8 nm), which could be of interest in 100-GHz-spaced WDM networks. From the view of extensive use of WDM technologies in high-speed optical communication networks [2628], a laudable goal would be to develop wavelength- and time-selective optical switching, dropping and swapping for 100-GHz-spaced WDM signals.

In this paper, we present single-PPLN-assisted optical switching, dropping and swapping for WDM signals [29]. Theoretical analyses with derived analytical solutions show the feasibility of narrow-band operation owing to the QPM condition of PPLN. By exploiting parametric depletion or combined parametric depletion and wavelength conversion in a PPLN waveguide, we demonstrate wavelength- and time-selective optical switching, dropping and swapping for 100-GHz-spaced 40-Gbit/s WDM signals.

2. Concept and principle

Figure 1 depicts the concept and operation principle of wavelength- and time-selective optical switching, dropping, and swapping. Second-order nonlinearities and their cascading in a single PPLN waveguide are employed to enable those optical signal processing functions.

 figure: Fig. 1

Fig. 1 Concept and principle of wavelength- and time-selective (a) switching, (b) dropping, and (c) swapping.

Download Full Size | PDF

For the switching operation, as shown in Fig. 1(a), the switching relies on the parametric depletion effect of the sum-frequency generation (SFG) in a single PPLN waveguide. A continuous-wave (CW) pump and a data-carrying signal participate in the SFG process, in which photons of the CW pump and data-carrying signal are annihilated to produce photons of the sum-frequency (SF) wave (not shown in Fig. 1(a)), resulting in the depletion of the data-carrying signal. Hence, the CW pump can switch the data-carrying signal “OFF” based on the parametric depletion effect of SFG. When employing multiple WDM channels (λS1,λS2, …, λSN), the wavelength channel to be switched/depleted (e.g., λS2) is determined by the pump (λP) with their wavelengths arranged to be nearly symmetric with respect to the quasi-phase matching (QPM) wavelength (λQPM) of PPLN. The narrow-band feature of the QPM condition enables wavelength-selective switching, i.e., only one of the WDM channels is switched without touching other channels. In addition, by using pulsed gate pump instead of CW pump, one can switch only the bits of data-carrying signal within the pump pulse duration. Consequently, it is possible to perform wavelength- and time-selective switching operation simply by adjusting the pump wavelength and choosing the pump pulse duration.

For the dropping operation, as shown in Fig. 1(b), a second pump (λP2) is adopted to access the switched information by cascaded second-order nonlinear interactions, i.e., cascaded sum- and difference-frequency generation (cSFG/DFG). Photons of data-carrying signal (λS2) and the first pump (λP1) are consumed in the SFG process to generate photons of an SF wave (not shown in Fig. 1(b)), which are simultaneously converted to photons of the second pump (λP2) and a newly generated idler (λi) in the subsequent DFG process. Thus the data-carrying signal is switched with its information copied onto a newly generated idler in the cSFG/DFG processes. The newly generated idler corresponds to a dropped copy of the switched signal with its wavelength decided by the second pump (1/λi=1/λS2+1/λP11/λP2). When employing multiple WDM channels (λS1,λS2, …, λSN), wavelength- and time-selective dropping is achievable by using pulsed gate pumps, changing pump wavelength, and varying pump pulse duration.

For the swapping operation, also known as data exchange/wavelength exchange/wavelength interchange, as shown in Fig. 1(c), two data-carrying signals (λS1, λS2) and two pumps (λP1, λP2) take part in the cSFG/DFG processes. Photons of the first data-carrying signal (λS1) and the first pump (λP1) are depleted to produce photons of SF wave (not shown in Fig. 1(c)), which are simultaneously transformed to photons of the second signal (λS2) and the second pump (λP2). Meanwhile, photons of the second data-carrying signal (λS2) and the second pump (λP2) are consumed to generate photons of SF wave, which are simultaneously converted to photons of the first signal (λS1) and the first pump (λP1). By appropriately controlling the power of two pumps, information swapping between two data-carrying signals (λS1,λS2), i.e., depletion of the first signal (λS1) and its conversion to the second signal (λS2) and vice versa, is available based on the combined effects of parametric depletion and wavelength conversion. When employing multiple WDM channels (λS1,λS2, …, λSN), it is possible to perform wavelength- and time-selective swapping by adopting pulsed gate pumps and adjusting the wavelength of two pumps. Only the swapping between two channels of interest (λS1,λS2) is enabled by setting the wavelength of two pumps (λP1, λP2) so that λS1 and λP1 as well as λS2 and λP2 are nearly symmetric related to the QPM wavelength of PPLN.

Remarkably, taking into account the narrow-band QPM condition, it holds the potential to implement wavelength- and time-selective optical switching, dropping and swapping for 100-GHz-spaced WDM signals.

3. Theory

Parametric depletion plays an important role in the proposed PPLN-based optical switching, dropping and swapping. We theoretically analyze the parametric depletion effect of the SFG process which enables optical switching. Generally speaking, the SFG-based optical switching involving one signal and one pump can be described using the coupled-mode equations under the slowly varying amplitude approximation written as

ASz+β1SASt+i2β2S2ASt2+12αSAS=iωSκSFGAP*ASFexp(iΔkSFGz)
APz+β1PAPt+i2β2P2APt2+12αPAP=iωPκSFGAS*ASFexp(iΔkSFGz)
ASFz+β1SFASFt+i2β2SF2ASFt2+12αSFASF=iωSFκSFGASAPexp(iΔkSFGz)
β1j=k/ω|ω=ωj=(njλjdn/dλ|λ=λj)/c,β2j=2k/ω2|ω=ωj=λj3/(2πc2)d2n/dλ2|λ=λj,j=S,P,SF
κSFG=deff2μ0/(cnSnPnSFAeff),ΔκSFG=2π(nSF/λSFnS/λSnP/λP1/Λ)
where AS, AP and ASF denote the normalized complex amplitudes of the signal, pump and sum-frequency wave, respectively. β1j and β2j are the first and second-order derivatives of the propagation constant kj with respect to the angular frequency ω, evaluated at ωj (j=S,P,SF). αj (j=S,P,SF) is the waveguide propagation loss. κSFG and ΔkSFG respectively refer to the coupling coefficient and phase mismatching for the second-order nonlinear interaction (i.e., sum-frequency generation). λj (j=S,P,SF) is the wavelength. deff=2d33/π is the effective nonlinear coefficient. μ0 is the permeability of free space. c is the light velocity in vacuum. nj (j=S,P,SF) is the refractive index. Aeff is the effective nonlinear interaction area. Λ is the poling period of PPLN waveguide.

In order to get analytical solutions, we simplify the coupled-mode equations (/t=0, 2/t2=0) and derive the following complex amplitude of the signal under the pump non-depletion approximation (AP(z)=AP(0)) and assuming a lossless waveguide (αj = 0).

AS(z)=AS(0)F(ΔkSFG,z)
F(ΔkSFG,z)=exp(iΔkSFGz/2)[cos(g1z)ig2sin(g1z)]
g1=1/2ΔkSFG2+4ωSωSFkSFG2|AP(0)|2,g2=ΔkSFG2/(ΔkSFG2+4ωSωSFkSFG2|AP(0)|2)
According to Eqs. (1)-(8), parametric depletion and SFG-based switching can be calculated.

Figures 2(a) -2(d) show depletion spectra of signal under different pump wavelengths of 1570.6, 1569.8, 1569.0 and 1568.2 nm, respectively. One can clearly see the narrow-band depletion spectra which are due to the QPM condition of the SFG process. The peak of depletion moves as changing the pump wavelength. Figure 3 presents detailed theoretical results for parametric depletion and SFG-based optical switching. As shown in Fig. 3(a), the depletion spectra can be shifted with the peak of depletion compatible with the ITU-grid by appropriately adjusting the pump wavelength. Shown in Fig. 3(b) is the depleted signal wavelength (peak of depletion) as a function of the pump wavelength. A linear relationship between the signal wavelength and pump wavelength is obtained and the signal wavelength decreases with the increase of pump wavelength. The bandwidth of depletion spectra is shown in Fig. 3(c). To achieve a high depletion larger than 20 dB, the bandwidth is calculated to be less than 0.06 nm under different pump wavelengths. Moreover, the sharp and narrow-band depletion spectra imply potential wavelength-selective operations among multiple WDM signals. As shown in Fig. 3(d), for a given pump wavelength, the bandwidth of depletion spectra is less than 100 GHz (0.8 nm) with signal depletion larger than 0.6 dB. The shadow regions shown in Fig. 3(d) mark the range of signal wavelength with negligible depletion less than 0.6 dB for different pump wavelengths. These narrow-band features together with negligible impacts on neighboring wavelength ranges offset from the peak of depletion can be clearly seen from the enlarged insets of depletion spectra shown in Fig. 2. Consequently, it is possible to implement wavelength-selective operations for 100-GHz-spaced WDM signals.

 figure: Fig. 2

Fig. 2 Theoretical results for the depletion spectra of signal under different pump wavelengths of (a) 1570.6 nm, (b) 1569.8 nm, (c) 1569.0 nm, and (d) 1568.2 nm. Insets show enlarged depletion spectra.

Download Full Size | PDF

 figure: Fig. 3

Fig. 3 Theoretical results for parametric depletion and SFG-based optical switching. (a) Depletion spectra under different pump wavelengths. (b) Depleted signal wavelength (peak of depletion) vs. pump wavelength. (c) Bandwidth of depletion spectra (>20 dB) vs. pump wavelength. (d) Signal wavelength (depletion = 0.6 dB) vs. pump wavelength. White areas (depletion > 0.6 dB) between two lines show a narrow bandwidth less than 100 GHz (0.8 nm). Shadow regions (depletion <0.6 dB) outside two lines indicate negligible impacts of parametric depletion and switching on neighboring channels.

Download Full Size | PDF

4. Experimental results and discussions

In the experiments, a 50-mm long PPLN waveguide fabricated by the reverse proton-exchange (RPE) technology is employed. It has a domain inversion period of 15 μm, an effective cross-section area of 50 μm2, a SFG QPM bandwidth of ~0.6 nm, a normalized conversion efficiency of 15%/mW for SFG, and propagation losses of 0.3 dB/cm in the 1.5 μm band and 0.6 dB/cm in the 0.77 μm band.

We first demonstrate wavelength- and time-selective optical switching for 100-GHz-spaced WDM signals. As shown in Fig. 4(a) , ITU-grid compatible four 40-Gbit/s 27-1 pseudo-random binary sequence (PRBS) WDM non-return-to-zero (NRZ) signals (S1: 1536.61 nm, S2: 1537.40 nm, S3: 1538.19 nm, S4: 1538.98 nm) are prepared with a channel spacing of 100 GHz. A synchronized pulsed gate pump is adopted with a pulse duration of ~1.2 ns and duty cycle of 3/127. Shown in Figs. 4(b)-4(e) and 4(f)-4(i) are recorded temporal waveforms and eye diagrams for four WDM NRZ signals (S1-S4).

 figure: Fig. 4

Fig. 4 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for four 40-Gbit/s WDM NRZ signals (S1: 1536.61 nm, S2: 1537.40 nm, S3: 1538.19 nm, S4: 1538.98 nm).

Download Full Size | PDF

Different WDM signals (S1-S4) can be switched by adjusting the pump wavelength. Figure 5(a) shows measured spectrum for wavelength-selective switching of S1 (1536.61 nm) with the pump wavelength tuned at 1570.6 nm. Shown in the inset is the recorded temporal waveform of the pulsed gate pump (pulse duration: ~1.2 ns, duty cycle: 3/127). Figures 5(b)-5(e) and 5(f)-5(i) depict measured temporal waveforms and eye diagrams of different signals after the switching of S1. One can clearly see only the bits of S1 within the pump pulse duration are switched off. Negligible impacts are observed on the other signals (S2, S3, S4), which is in agreement with theoretical analyses.

 figure: Fig. 5

Fig. 5 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S1 (1536.61 nm) with the pump wavelength tuned at 1570.6 nm.

Download Full Size | PDF

When using the same pulsed gate pump (pulse duration: ~1.2 ns, duty cycle: 3/127) while tuning the pump wavelength at 1569.8 nm, wavelength-selective switching of S2 (1537.40 nm) is achieved. Figure 6(a) shows measured spectrum for the switching of S2. The inset of Fig. 6(a) depicts temporal waveform of pulsed gate pump. Figures 6(b)-6(e) display recorded temporal waveforms for four signals (S1-S4). It is shown that the bits of S2 within the pump pulse duration are switched off while other signals (S1, S3, S4) are not touched. The eye diagrams of different signals are shown in Figs. 6(f)-6(i).

 figure: Fig. 6

Fig. 6 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S2 (1537.40 nm) with the pump wavelength tuned at 1569.8 nm.

Download Full Size | PDF

Figure 7(a) shows measured spectrum for wavelength-selective switching of S3 (1538.19 nm) by tuning the pump wavelength at 1569.0 nm. The inset of Fig. 7(a) depicts temporal waveform of the pulsed gate pump (pulse duration: ~1.2 ns, duty cycle: 3/127). By observing the temporal waveforms of different signals (S1-S4) as shown in Figs. 7(b)-7(e), one can see that the bits of S3 which feel the pulsed gate pump are switched off due to the parametric depletion effect of SFG process. It is also noted that the bits of S3 out of the pump pulse duration and other signals (S1, S2, S4) are not touched. Figures 7(f)-7(i) present recorded eye diagrams of different signals (S1-S4) for the switching of S3.

 figure: Fig. 7

Fig. 7 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S3 (1538.19 nm) with the pump wavelength tuned at 1569.0 nm.

Download Full Size | PDF

When the pulsed gate pump (pulse duration: ~1.2 ns, duty cycle: 3/127) is tuned at 1568.2 nm, we obtain the wavelength-selective switching of S4 (1538.98 nm). The measured spectrum is depicted in Fig. 8(a) with the inset showing the temporal waveform of pulsed gate pump. The wavelength-selective switching of S4 without impacting on other signals (S1-S3) is confirmed from the measured temporal waveforms of different signals (S1-S4) as shown in Figs. 8(b)-8(e). Only the bits of S4 within the pump pulse duration are switched off. Shown in Figs. 8(f)-8(i) are recorded eye diagrams of different signals (S1-S4).

 figure: Fig. 8

Fig. 8 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S4 (1538.98 nm) with the pump wavelength tuned at 1568.2 nm.

Download Full Size | PDF

Flexible time-selective switching is verified by changing the pump pulse duration. Figure 9(a) displays the temporal waveform of pulsed gate pump (1568.2 nm) which has two pulse durations of 1.2 and 0.4 ns with an interval of 0.4 ns. When the pump is off, the temporal waveform of S4 is shown in Fig. 9(b). When the pump is on, as shown in Fig. 9(c), the bits of S4 within the two pump pulse durations are switched off. Figures 9(d)-9(f) depict recorded temporal waveforms of S1-S3 for the switching of S4. Negligible impacts on S1-S4 are observed. Figures 9(g)-9(k) show measured eye diagrams of different signals for the time-selective switching of S4.

 figure: Fig. 9

Fig. 9 Measured (a)-(f) temporal waveforms and (g)-(k) eye diagrams for the time-selective switching of S4 (1538.98 nm) with the pump wavelength tuned at 1568.2 nm.

Download Full Size | PDF

Figure 10 shows another example of flexible time-selective switching. As shown in Fig. 10(a), the pulsed gate pump has three pulse durations of 0.6, 0.5 and 0.2 ns with intervals of 0.3 and 0.2 ns. Figures 10(b)-10(f) display measured temporal waveforms of different signals, from which one can clearly see the time-selective switching of S2 within three pump pulse durations without affecting other signals (S1, S3, S4). The recorded eye diagrams of different signals are shown in Figs. 10(g)-10(k).

 figure: Fig. 10

Fig. 10 Measured (a)-(f) temporal waveforms and (g)-(k) eye diagrams for the time-selective switching of S2 (1537.40 nm) with the pump wavelength tuned at 1569.8 nm.

Download Full Size | PDF

The achieved results shown in Figs. 4-10 confirm the successful realization of wavelength- and time-selective optical switching by tuning the pump wavelength and changing the pump pulse duration. Figures 11(a) -11(c) plot measured bit-error rate (BER) performance for the wavelength- and time-selective optical switching corresponding to Figs. 5-8, Fig. 9 and Fig. 10, respectively. The power penalties are measured to be less than 1 dB at a BER of 10−9.

 figure: Fig. 11

Fig. 11 Measured BER curves for wavelength- and time-selective switching. (a)-(c) correspond to Figs. 5-8, 9, 10.

Download Full Size | PDF

We then demonstrate wavelength- and time-selective optical dropping for 100-GHz-spaced WDM signals (S1: 1536.61 nm, S2: 1537.40 nm, S3: 1538.19 nm, S4: 1538.98 nm), as shown in Fig. 12 . Pump1 (λP1) is tuned at 1568.2 nm to switch S4 with its data information copied onto an idler (dropping) by the combined parametric depletion and wavelength conversion of cSFG/DFG when pump2 (λP2) is employed. By tuning pump2 from 1566.6 to 1539.5 nm, the converted idler is varied from 1540.1 to 1567.2 nm with an average conversion efficiency larger than −22 dB and a fluctuation of conversion efficiency less than 1.1 dB, as shown in Figs. 12(a)-12(c). Shown in Figs. 12(d)-12(i) are measured temporal waveforms with pump2 tuned at 1557.6 nm and idler generated at 1548.9 nm. One can clearly see the switching of S4 within the pump pulse duration and its dropping onto the idler. It is also noted that other signals (S1, S2, S3) are not touched due to the narrow-band QPM condition. Figures 12(j)-12(n) depict recorded eye diagrams of different signals (S1-S4) for the wavelength- and time-selective dropping of S4.

 figure: Fig. 12

Fig. 12 Measured results for wavelength- and time-selective dropping. (a)(b) Spectra. (c) Conversion efficiency. (d)-(i) Temporal waveforms. (j)-(n) Eye diagrams. (d)-(n) Wavelength- and time-selective switching and dropping of S4 (1538.98 nm) with the pump1 and pump2 wavelengths tuned at 1568.2 and 1557.6 nm and the idler generated at 1548.9 nm.

Download Full Size | PDF

As shown in Fig. 13 , by changing the pump1 wavelength to be 1570.6, 1569.8 and 1569.0 nm, we also achieve wavelength-selective switching and dropping of S1, S2 and S3, respectively.

 figure: Fig. 13

Fig. 13 Measured temporal waveforms for wavelength-selective switching and dropping. (a)(b) S1 (1536.61 nm). (c)(d) S2 (1537.40 nm). (e)(f) S3 (1538.19 nm). The pump1 wavelength is tuned to be 1570.6 nm in (a)(b), 1569.8 nm in (c)(d), and 1569.0 nm in (e)(f).

Download Full Size | PDF

We finally demonstrate optical swapping between two 100-GHz-spaced 40-Gbit/s signals (SA: 1542.54 nm, SB: 1543.33 nm). As shown in Fig. 14(a) , two pulsed gate pumps (pulse duration: ~1.2 ns, duty cycle: 3/127) are adjusted at 1564.0 (PA) and 1562.9 nm (PB) so that SA and PA as well as SB and PB are approximately symmetric with respect to the QPM wavelength of PPLN (~1553 nm). In order to verify the data swapping operation, we record temporal waveforms in Figs. 14(b)-14(e) and eye diagrams in Figs. 14(f)-14(i). One can clearly see that the bits within the pump pulse duration are successfully swapped between SA and SB. The distorted signal quality after data swapping can be ascribed to the beating effect between the newly converted signal (wavelength conversion) and the original residual signal (incomplete parametric depletion). Additionally, wavelength- and time-selective data swapping for WDM signals is also available due to narrow-band QPM condition by appropriately tuning pump wavelength and choosing pump pulse duration [21].

 figure: Fig. 14

Fig. 14 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for optical swapping between two 100-GHz-spaced signals (SA: 1542.54 nm, SB: 1543.33 nm).

Download Full Size | PDF

The obtained results shown in Figs. 4-14 indicate the successful implementation of single-PPLN-assisted wavelength- and time-selective optical switching, dropping and swapping for 100-GHz-spaced WDM signals. With future improvement, enabled by the appropriate design and optimization of the parameters of PPLN (e.g., length, domain inversion period, etc.) for QPM engineering, it holds a potential to further extend the proposed wavelength- and time-selective switching, dropping and swapping to be available for WDM signals with even narrower channel spacing, i.e. 50 GHz and 25 GHz.

5. Conclusion

In summary, by exploiting second-order nonlinearities and their cascading in a PPLN waveguide, we have proposed a scheme to realize wavelength- and time-selective optical switching, dropping and swapping. We have theoretically analyzed the parametric depletion effect and its narrow-band feature due to QPM condition of PPLN from the derived analytical solutions. Based on parametric depletion of SFG, wavelength- and time-selective optical switching for ITU-grid compatible 100-GHz-spaced 40-Gbit/s WDM signals has been implemented. Based on combined parametric depletion and wavelength conversion of cSFG/DFG, wavelength- and time-selective optical dropping for 100-GHz-spaced 40-Gbit/s WDM signals and optical swapping between two 100-GHz-spaced 40-Gbit/s signals have been demonstrated. The proposed single-PPLN-assisted wavelength- and time-selective optical switching, dropping and swapping might be further extended to WDM signals with narrower channel spacing (e.g., 50 GHz, 25 GHz) by optimizing the parameters of PPLN.

Acknowledgments

This work was supported by the National Natural Science Foundation of China (NSFC) under grants 61077051, 61222502, 11274131, the Program for New Century Excellent Talents in University (NCET-11-0182), the Natural Science Foundation of Hubei Province of China under Grant 2011CDB032, and the Huawei Innovation Research Program (YJCB2011061RE).

References and links

1. A. E. Willner, O. F. Yilmaz, J. Wang, X. Wu, A. Bogoni, L. Zhang, and S. R. Nuccio, “Optically efficient nonlinear signal processing,” IEEE J. Sel. Top. Quantum Electron. 17(2), 320–332 (2011). [CrossRef]  

2. S. Radic, “Parametric signal processing,” IEEE J. Sel. Top. Quantum Electron. 18(2), 670–680 (2012). [CrossRef]  

3. K. E. Stubkjaer, “Semiconductor optical amplifier-based all-optical gates for high-speed optical signal processing,” IEEE J. Sel. Top. Quantum Electron. 6(6), 1428–1435 (2000). [CrossRef]  

4. J. Wang, J. Sun, C. Luo, and Q. Sun, “Flexible all-optical wavelength conversions of 1.57-ps pulses exploiting cascaded sum- and difference frequency generation (cSFG/DFG) in a PPLN waveguide,” Appl. Phys. B 83(4), 543–548 (2006). [CrossRef]  

5. C. Langrock, S. Kumar, J. E. McGeehan, A. E. Willner, and M. M. Fejer, “All-optical signal processing using nonlinearities in guided-wave devices,” J. Lightwave Technol. 24(7), 2579–2592 (2006). [CrossRef]  

6. J. Wang, J. Sun, and Q. Sun, “Proposal for all-optical format conversion based on a periodically poled lithium niobate loop mirror,” Opt. Lett. 32(11), 1477–1479 (2007). [CrossRef]   [PubMed]  

7. J. Wang, J. Sun, Q. Sun, D. Wang, M. Zhou, X. Zhang, D. Huang, and M. M. Fejer, “Experimental observation of all-optical non-return-to-zero-to-return-to-zero format conversion based on cascaded second-order nonlinearity assisted by active mode-locking,” Opt. Lett. 32(16), 2462–2464 (2007). [CrossRef]   [PubMed]  

8. B. J. Eggleton, T. D. Vo, R. Pant, J. Schr, M. D. Pelusi, D. Yong Choi, S. J. Madden, and B. Luther-Davies, “Photonic chip based ultrafast optical processing based on high nonlinearity dispersion engineered chalcogenide waveguides,” Laser Photon. Rev. 6(1), 97–114 (2012). [CrossRef]  

9. L. K. Oxenlowe, M. Hua Ji, Galili, Minhao Pu, H. C. H. Hao Hu, K. Mulvad, J. M. Yvind, A. T. Hvam, Clausen, and P. Jeppesen, “Silicon photonics for signal processing of Tbit/s serial data signals,” IEEE J. Sel. Top. Quantum Electron. 18(2), 996–1005 (2012). [CrossRef]  

10. J. Wang, H. Huang, X. Wang, J. Y. Yang, and A. E. Willner, “Reconfigurable 2.3-Tbit/s DQPSK simultaneous add/drop, data exchange and equalization using double-pass LCoS and bidirectional HNLF,” Opt. Express 19(19), 18246–18252 (2011). [CrossRef]   [PubMed]  

11. K. Uesaka, K. K. Y. Wong, M. E. Marhic, and L. G. Kazovsky, “Wavelength exchange in a highly nonlinear dispersion-shifted fiber: theory and experiments,” IEEE J. Sel. Top. Quantum Electron. 8(3), 560–568 (2002). [CrossRef]  

12. R. W. L. Fung, H. K. Y. Cheung, and K. K. Y. Wong, “Widely tunable wavelength exchange in anomalous-dispersion regime,” IEEE Photon. Technol. Lett. 19(22), 1846–1848 (2007). [CrossRef]  

13. J. Wang, Z. Bakhtiari, S. R. Nuccio, O. F. Yilmaz, X. Wu, and A. E. Willner, “Phase-transparent optical data exchange of 40 Gbit/s differential phase-shift keying signals,” Opt. Lett. 35(17), 2979–2981 (2010). [CrossRef]   [PubMed]  

14. M. Shen, X. Xu, T. I. Yuk, and K. K. Y. Wong, “Byte-level parametric wavelength exchange for narrow pulsewidth return-to-zero signals,” IEEE Photon. Technol. Lett. 21(21), 1591–1593 (2009). [CrossRef]  

15. V. Van, T. A. Ibrahim, K. Ritter, P. P. Absil, F. G. Johnson, R. Grover, J. Goldhar, and P.-T. Ho, “All-optical nonlinear switching in GaAs-AlGaAs microring resonators,” IEEE Photon. Technol. Lett. 14(1), 74–76 (2002). [CrossRef]  

16. V. R. Almeida, C. A. Barrios, R. R. Panepucci, and M. Lipson, “All-optical control of light on a silicon chip,” Nature 431(7012), 1081–1084 (2004). [CrossRef]   [PubMed]  

17. K. R. Parameswaran, M. Fujimura, M. H. Chou, and M. M. Fejer, “Low-power all-optical gate based on sum frequency mixing in APE waveguides in PPLN,” IEEE Photon. Technol. Lett. 12(6), 654–656 (2000). [CrossRef]  

18. T. Suhara and H. Ishizuki, “Integrated QPM sum-frequency generation interferometer device for ultrafast optical switching,” IEEE Photon. Technol. Lett. 13(11), 1203–1205 (2001). [CrossRef]  

19. Y. L. Lee, H. Suche, Y. H. Min, J. H. Lee, W. Grundkötter, V. Quiring, and W. Sohler, “Wavelength- and time-selective all-optical channel dropping in periodically poled Ti:LiNbO3 channel waveguides,” IEEE Photon. Technol. Lett. 15(7), 978–980 (2003). [CrossRef]  

20. J. Wang and Q. Z. Sun, “Theoretical analysis of power swapping in quadratic nonlinear medium,” Appl. Phys. Lett. 96(8), 081108 (2010). [CrossRef]  

21. J. Wang, S. R. Nuccio, X. Wu, O. F. Yilmaz, L. Zhang, I. Fazal, J. Y. Yang, Y. Yue, and A. E. Willner, “40 Gbit/s optical data exchange between wavelength-division-multiplexed channels using a periodically poled lithium niobate waveguide,” Opt. Lett. 35(7), 1067–1069 (2010). [CrossRef]   [PubMed]  

22. A. Bogoni, X. Wu, S. R. Nuccio, J. Wang, Z. Bakhtiari, and A. E. Willner, “Photonic 640-Gb/s reconfigurable OTDM add-drop multiplexer based on pump depletion in a single PPLN waveguide,” IEEE J. Sel. Top. Quantum Electron. 18(2), 709–716 (2012). [CrossRef]  

23. T. K. Liang, L. R. Nunes, T. Sakamoto, K. Sasagawa, T. Kawanishi, M. Tsuchiya, G. R. A. Priem, D. Van Thourhout, P. Dumon, R. Baets, and H. K. Tsang, “Ultrafast all-optical switching by cross-absorption modulation in silicon wire waveguides,” Opt. Express 13(19), 7298–7303 (2005). [CrossRef]   [PubMed]  

24. L. Yin, J. Zhang, P. M. Fauchet, and G. P. Agrawal, “Optical switching using nonlinear polarization rotation inside silicon waveguides,” Opt. Lett. 34(4), 476–478 (2009). [CrossRef]   [PubMed]  

25. S. Abaslou and V. Ahmadi, “Compact all-optical switch for WDM networks based on Raman effect in silicon nanowavegide,” Opt. Lett. 37(1), 40–42 (2012). [CrossRef]   [PubMed]  

26. Y. Miyamoto, T. Kataoka, K. Yonenaga, M. Tomizawa, A. Hirano, S. Kuwahara, and Y. Tada, “WDM field trials of 43-Gb/s/channel transport system for optical transport network,” J. Lightwave Technol. 20(12), 2115–2128 (2002). [CrossRef]  

27. M. Sauer, J. E. Hurley, S. Ten, J. J. Ferner, and S. P. Colby, “1.6 Tbit/s transmission over 2160 km of field-deployed dispersion-managed fibre without per channel dispersion compensation,” Electron. Lett. 39(9), 728–730 (2003). [CrossRef]  

28. A. H. Gnauck, R. W. Tkach, A. R. Chraplyvy, and T. Li, “High-capacity optical transmission systems,” J. Lightwave Technol. 26(9), 1032–1045 (2008). [CrossRef]  

29. J. Wang, H. Fu, D. Geng, and A. Willner, “All-optical wavelength-/time-selective switching/dropping/swapping for 100-GHz-spaced WDM signals using a periodically poled lithium niobate waveguide,” in European Conference and Exhibition on Optical Communication, OSA Technical Digest (online) (Optical Society of America, 2012), paper Th.1.A.5.

Cited By

Optica participates in Crossref's Cited-By Linking service. Citing articles from Optica Publishing Group journals and other participating publishers are listed here.

Alert me when this article is cited.


Figures (14)

Fig. 1
Fig. 1 Concept and principle of wavelength- and time-selective (a) switching, (b) dropping, and (c) swapping.
Fig. 2
Fig. 2 Theoretical results for the depletion spectra of signal under different pump wavelengths of (a) 1570.6 nm, (b) 1569.8 nm, (c) 1569.0 nm, and (d) 1568.2 nm. Insets show enlarged depletion spectra.
Fig. 3
Fig. 3 Theoretical results for parametric depletion and SFG-based optical switching. (a) Depletion spectra under different pump wavelengths. (b) Depleted signal wavelength (peak of depletion) vs. pump wavelength. (c) Bandwidth of depletion spectra (>20 dB) vs. pump wavelength. (d) Signal wavelength (depletion = 0.6 dB) vs. pump wavelength. White areas (depletion > 0.6 dB) between two lines show a narrow bandwidth less than 100 GHz (0.8 nm). Shadow regions (depletion <0.6 dB) outside two lines indicate negligible impacts of parametric depletion and switching on neighboring channels.
Fig. 4
Fig. 4 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for four 40-Gbit/s WDM NRZ signals (S1: 1536.61 nm, S2: 1537.40 nm, S3: 1538.19 nm, S4: 1538.98 nm).
Fig. 5
Fig. 5 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S1 (1536.61 nm) with the pump wavelength tuned at 1570.6 nm.
Fig. 6
Fig. 6 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S2 (1537.40 nm) with the pump wavelength tuned at 1569.8 nm.
Fig. 7
Fig. 7 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S3 (1538.19 nm) with the pump wavelength tuned at 1569.0 nm.
Fig. 8
Fig. 8 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for wavelength-selective switching of S4 (1538.98 nm) with the pump wavelength tuned at 1568.2 nm.
Fig. 9
Fig. 9 Measured (a)-(f) temporal waveforms and (g)-(k) eye diagrams for the time-selective switching of S4 (1538.98 nm) with the pump wavelength tuned at 1568.2 nm.
Fig. 10
Fig. 10 Measured (a)-(f) temporal waveforms and (g)-(k) eye diagrams for the time-selective switching of S2 (1537.40 nm) with the pump wavelength tuned at 1569.8 nm.
Fig. 11
Fig. 11 Measured BER curves for wavelength- and time-selective switching. (a)-(c) correspond to Figs. 5-8, 9, 10.
Fig. 12
Fig. 12 Measured results for wavelength- and time-selective dropping. (a)(b) Spectra. (c) Conversion efficiency. (d)-(i) Temporal waveforms. (j)-(n) Eye diagrams. (d)-(n) Wavelength- and time-selective switching and dropping of S4 (1538.98 nm) with the pump1 and pump2 wavelengths tuned at 1568.2 and 1557.6 nm and the idler generated at 1548.9 nm.
Fig. 13
Fig. 13 Measured temporal waveforms for wavelength-selective switching and dropping. (a)(b) S1 (1536.61 nm). (c)(d) S2 (1537.40 nm). (e)(f) S3 (1538.19 nm). The pump1 wavelength is tuned to be 1570.6 nm in (a)(b), 1569.8 nm in (c)(d), and 1569.0 nm in (e)(f).
Fig. 14
Fig. 14 Measured (a) spectrum, (b)-(e) temporal waveforms, and (f)-(i) eye diagrams for optical swapping between two 100-GHz-spaced signals (SA: 1542.54 nm, SB: 1543.33 nm).

Equations (8)

Equations on this page are rendered with MathJax. Learn more.

A S z + β 1S A S t + i 2 β 2S 2 A S t 2 + 1 2 α S A S =i ω S κ SFG A P * A SF exp(iΔ k SFG z)
A P z + β 1P A P t + i 2 β 2P 2 A P t 2 + 1 2 α P A P =i ω P κ SFG A S * A SF exp(iΔ k SFG z)
A SF z + β 1SF A SF t + i 2 β 2SF 2 A SF t 2 + 1 2 α SF A SF =i ω SF κ SFG A S A P exp(iΔ k SFG z)
β 1j = k/ω | ω= ω j =( n j λ j dn/dλ | λ= λ j )/c, β 2j = 2 k/ ω 2 | ω= ω j = λ j 3 /(2π c 2 ) d 2 n/d λ 2 | λ= λ j ,j=S,P,SF
κ SFG = d eff 2 μ 0 /(c n S n P n SF A eff ) , Δ κ SFG =2π( n SF / λ SF n S / λ S n P / λ P 1/Λ)
A S (z)= A S (0)F(Δ k SFG ,z)
F(Δ k SFG ,z)=exp(iΔ k SFG z/2)[cos( g 1 z)i g 2 sin( g 1 z)]
g 1 =1/2 Δ k SFG 2 +4 ω S ω SF k SFG 2 | A P (0) | 2 , g 2 = Δ k SFG 2 /(Δ k SFG 2 +4 ω S ω SF k SFG 2 | A P (0) | 2 )
Select as filters


Select Topics Cancel
© Copyright 2024 | Optica Publishing Group. All rights reserved, including rights for text and data mining and training of artificial technologies or similar technologies.