Open Access
Add to BibSonomy
Abstract
All-optical wavelength conversion for 2×11.64 GBaud adaptively-modulated orthogonal frequency division multiplexing (AM-OFDM) signals with QPSK/16QAM formats is experimentally demonstrated in a silicon waveguide. The AM-OFDM signal with partly higher- (and lower-) order formats on lower- (and higher-) frequency subcarriers has better overall conversion performance in receiving optical signal-to-noise ratio and power penalty. In comparison with the OFDM-QPSK signal, at the BER of 3.8×10−3, the bit rate increases 11.64 Gbit/s per channel almost without conversion power penalty increased by replacing the QPSK sequence with the 16QAM sequence on half subcarriers.
© 2017 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
All-optical wavelength conversion is used to avoid network traffics in wavelength-routing optical communication networks and has been demonstrated in highly-nonlinear fibers, periodically-poled lithium niobate crystals, semiconductor optical amplifiers, and silicon waveguides [1–3]. In particular, silicon waveguides have been considered as a promising way to realize nonlinear optical devices due to their strong light confinement and highly nonlinear response. By using four-wave mixing (FWM) in silicon waveguides, all-optical wavelength conversion has been well realized for binary and multi-level modulation formats, such as quadrature phase shift-keying (QPSK) [2] and quadrature amplitude modulation (QAM) [3], which can effectively improve the transmission capacity thanks to their high spectral efficiency [4]. Furthermore, multiplexing is another efficient way to improve the capacity by extending the channels number, such as wavelength-division multiplexing (WDM), time-division multiplexing (TDM), and polarization-division multiplexing (PDM). FWM-based wavelength conversion in silicon waveguides is available to these multiplexing data [5, 6]. Although the channel number can be increased using these multiplexing technologies, the single-channel data is still modulated on a single carrier. Orthogonal frequency division multiplexing (OFDM) provides a new way to combine the modulation and multiplexing together [7], which has been widely used in radio-frequency fields [8, 9]. OFDM employs multiple carriers to convey the information from the source to the destination within the allocated bandwidth and transfers the data stream to many lower-rate subcarrier tones. The technology can be also used in optical domain. For both discrete multi-tone (DMT) and coherent detected systems, optical OFDM perfectly inherits the advantages of the wireless OFDM, such as simple one-tap frequency domain equalization. Compared with optical DMT, coherent optical OFDM can offer higher spectral efficiency and higher receiver sensitivity. The transmission of optical OFDM signals has been realized in multi-band [10], dual-polarization [11], and WDM systems [12, 13]. Also, direct-detection and coherent optical fast-OFDM have been proposed to further improve the spectral efficiency by reducing subcarrier spacing [14]. For the OFDM signals with identical m-QAM [15] modulated subcarriers, integrated wavelength conversion has been realized using silicon waveguides.
For these OFDM signals with an identical coherent modulation format on all the subcarriers, the transmission error probability increases for the higher-frequency subcarriers because they have lower signal-to-noise ratios (SNRs) due to the roll-off of frequency response of electrical and optoelectronic components [16]. In order to balance the SNRs of these subcarriers, lower- (or higher-) level modulation formats can be loaded on higher- (or lower-) frequency subcarriers to keep the data transmission quality since lower (or higher) SNRs are required [17, 18]. The transmission performance of such adaptively-modulated OFDM (AM-OFDM) signals have been numerically analyzed, which shows the robustness of transmission capacity and network flexibility [19, 20]. In this paper, we propose and experimentally demonstrate, for the first time to our best knowledge, the all-optical wavelength conversion in a silicon waveguide for 2 × 11.64 GBaud WDM AM-OFDM signals, where QPSK and 16QAM modulation formats are partly loaded on the 64 subcarriers of the OFDM signal. Comparing with the identical-modulated OFDM signals, the overall performance of bit rate, receiving sensitivity, and conversion power penalty is greatly improved.
2. Principle
FWM has the parallel processing ability of multiple WDM channels, and all the phase and amplitude information can be strictly preserved. Therefore, FWM in silicon waveguides can support the wavelength conversion of WDM AM-OFDM signals. When multiple-channel WDM AM-OFDM signals and a high-power pump P are injected into a silicon waveguide, multiple degenerate FWM processes will occur simultaneously between every WDM channel and the pump. In this way, multiple converted idlers will be generated, as depicted in Fig. 1(a). The phase relationship among the pump, signal, and idler is (). If the pump is provided by a continuous-wave (CW) light, whose phase can be supposed to be for QPSK/16QAM-modulated OFDM demodulation, the idler’s phase becomes (). Therefore, the WDM AM-OFDM signals can be converted to the idlers.
Fig. 1 (a) Principle of the wavelength conversion for WDM-compatible OFDM signals. (b) Schematic diagrams for four modulated schemes across all 64 subcarriers.
The arrangement of the OFDM subcarriers should follow the IEEE 802.11 specification. Taking 64 data subcarriers as an example, they are numbered from −32, −31, …,−2, −1 and 1, 2, …, 31, 32. The smaller numbers (±1, ±2, …) correspond to lower-frequency subcarriers and the larger numbers (± 32, ± 31, …) correspond to higher-frequency subcarriers. Different subcarriers are allowed loading different modulation formats independently. As shown in Fig. 1(b), four modulation schemes are used to analyze the wavelength conversion performance of the WDM AM-OFDM signals. For case A (or case D), all 64 subcarriers are modulated with QPSK (or 16QAM) symbols, denoted as OFDM-QPSK (or OFDM-16QAM). For case B, the 16 lower-frequency subcarriers (1 to 8 and −1 to −8, 1/4 of the subcarriers) are modulated with 16QAM sequences, and the rest 48 subcarriers (3/4 of the subcarriers) are modulated with QPSK sequences, denoted as OFDM-3/4QPSK + 1/4QAM. While for case C, the OFDM signal is modulated by QPSK symbols on 32 higher-frequency subcarriers and 16QAM symbols on the other 32 lower-frequency subcarriers, denoted as OFDM-1/2QPSK+1/2QAM.
3. Experimental setup
Figure 2(a) shows the experimental setup of the wavelength conversion for 2×11.64 GBaud WDM AM-OFDM signals. Two CW optical carriers with the wavelengths of 1550.92 and 1551.70 nm, emitted form a distributed feedback (DFB) laser (Koheras Bootik E15 with a linewidth <50 kHz) and a tunable laser (Agilent N7714A with a linewidth of ~100 kHz), are modulated by multi-level electrical OFDM sequences from an arbitrary waveform generator (AWG, Tektronix 70002A, 25 GS/s sampling rate) via an in-phase quadrature modulator (IQM). Here, the AM-OFDM signals with 72 data subcarriers are generated digitally, in which 64 subcarriers are used to carry the payloads and the other 8 subcarriers are selected as the cyclic prefix (CP) to eliminate the inter-symbol interference (ISI). An inverse fast Fourier transform (IFFT) with a size of 128 is used to convert the signal to time domain. 11 training symbols for every 1023 payload symbols are employed for the channel estimation. The bit rates of the two-channel WDM-OFDM signals are calculated to be 46.56, 58.20, 69.84, and 93.12 Gbit/s for cases A-D, respectively. An amplified spontaneous emission (ASE) noise source is introduced through a multiplexer (MUX1) to tune the optical SNR (OSNR) of the AM-OFDM signals S1 and S2. They are separated using a de-multiplexer (DMUX) and decorrelated using a 550-ps optical delay line (ODL). After being amplified to 22.59 and 21.65 dBm, they are combined with a CW pump P provided by another tunable laser (Agilent N7714A) via a multiplexer (MUX2), whose bandwidth is 0.8 nm to suppress the amplifier noises. The pump is at 1549.36 nm with a power of 30.7 dBm. And then, the pump and signals are launched into a 17-mm-long silicon waveguide with an effective mode area of about 5 μm2. According to its dispersion profile [21], the conversion bandwidth is calculated to be about 32 nm, which is enough to support the C-band wavelength conversion. The coupling loss for each fiber-to-waveguide is about 1.5 dB, and the propagation loss in the waveguide is about 1.7 dB. Their polarizations are tuned to be parallel by the polarization controllers (PC3-PC5) and the polarization beam coupler (PBC) and aligned to the transverse electric (TE) or transverse magnetic (TM) mode of the silicon waveguide by PC6 to achieve the maximum FWM efficiencies. In the waveguide, the FWM processes occur between the pump P and the signal S1 (or S2) to generate the idler I1 (or I2), as shown by the optical spectrum in Fig. 2(b), which carries the same OFDM sequences as the input signals. The FWM efficiencies are calculated to be −38.0 and −39.5 dB, respectively. After filtering and pre-amplifying, the maximum OSNR of each idler is about 20 dB. The idler is separately demodulated via coherent detection [2, 3], with a local oscillator (Agilent 81940), a four-channel digital storage oscilloscope (Tektronix 72004C), and digital signal processing (DSP).
Fig. 2 (a) Experimental setup of the wavelength conversion for the two-channel WDM AM-OFDM signals in a silicon waveguide; (b) Measured optical spectrum after the silicon waveguide.
4. Results and discussion
Figure 3 shows the constellation diagrams of the two-channel WDM AM-OFDM signals and their converted idlers for the four modulation cases. Their root mean square (RMS) error vector magnitudes (EVMs) are calculated, as also shown in Fig. 3. Compared with the input AM-OFDM signals, the EVMs of the converted idlers only increase a little (<2.2%), which indicates that the silicon-based wavelength conversion has good performance. The single degradation is mainly caused by the noises including from the amplified spontaneous emission (ASE) of the amplifiers and the fluctuation of laser sources.
Fig. 3 Constellation diagrams for the AM-OFDM signals and the converted idlers with four different modulated schemes. (a1)-(d1) OFDM-QPSK; (a2)-(d2) OFDM-3/4QPSK+1/4QAM; (a3)-(d3) OFDM-1/2QPSK+1/2QAM; (a4)-(d4) OFDM-QAM.
Figure 4 shows the measured BERs of each subcarrier for cases A-D at the OSNR of ~21 dB. When all the 64 subcarriers are modulated with QPSK data (case A) or only the 16 lower-frequency subcarriers are replaced by 16QAM data (case B), the measured BERs of the signals are almost zero. As the modulation formats is replaced by 16QAM data for the 32 higher-frequency subcarriers (case C), the BERs increase a little, but they are still below 4×10−4. When all the 64 subcarriers are modulated with 16QAM data, the BERs rapidly increase (more than 1×10−3) for the higher-frequency 32 subcarriers (from 17 to 32 and −17 to −32). Therefore, AM-OFDM signal shows a significant improvement of the overall performance of wavelength conversion on signal quality and bit rate by partly using higher-order (and lower-order) modulation formats on the lower-frequency (and higher-frequency) subcarriers. As in case D, one can find that different subcarriers have different BERs under the same modulation format since the SNRs differ for the subcarriers at different frequencies. Therefore, the system capacity can be further enhanced by bit-loading or bit-and-power loading.
Fig. 4 Measured BER results for 64 subcarriers with four different modulation cases at the OSNR of ~21 dB.
The BER results for the two-channel WDM AM-OFDM signals and the converted idlers are measured as the signal OSNR varies, as shown in Fig. 5. The required OSNRs and the calculated power penalties at the BER of 3.8×10−3 are shown in Table 1. Although the bit rate increases from case A to case C by introducing higher-order 16QAM formats onto the lower-frequency subcarriers, the conversion power penalty increases very slightly (i.e., ~0.4 dB). The transmission rates of cases B and C increase by 11.64 and 23.28 Gbit/s compared to case A, respectively. However, if the higher-order 16QAM formats are used to all the higher-frequency subcarriers, as case D, the power penalty significantly increases by ~2.7 dB). Therefore, the AM-OFDM technology by partly modulating the lower-frequency subcarriers with higher-order modulation formats is an effective way to increase the bite rate without obviously degrading the converted signal quality, and it is also compatible to WDM technology.
Fig. 5 Measured BER results of the WDM signals and converted idlers for (a) OFDM-QPSK, (b) OFDM-3/4QPSK + 1/4QAM, (c) OFDM-1/2QPSK + 1/2QAM, and (d) OFDM-QAM signals.
Table 1. Calculated OSNRs and power penalties at the BER of 3.8 × 10−3.
5. Conclusion
All-optical wavelength conversion for 2 × 11.64 GBaud WDM AM-OFDM signals has been demonstrated using FWM in a silicon waveguide. At the BER of 3.8×10−3, the bit rate increases by 33.3% (i.e., 11.64 Gbit/s) almost without conversion power penalty increased by replacing the QPSK sequence with the 16QAM sequence on half subcarriers (lower-frequency subcarriers). Although the bit rate can be further increased by replacing the QPSK sequences by the 16QAM sequences on the rest higher-frequency subcarriers, the power penalty increases obviously. The AM-OFDM signal with partly higher- (and lower-) order formats on lower- (and higher-) frequency subcarriers has better overall conversion performance.
Funding
National Natural Science Foundation of China (61475138 and 61675177).
References and links
1. J. Yamawaku, H. Takara, T. Ohara, K. Sato, A. Takada, T. Morioka, O. Tadanaga, H. Miyazawa, and M. Asobe, “Simultaneous 25 GHz-spaced DWDM wavelength conversion of 1.03 Tbit/s (103/spl times/10 Gbit/s) signals in PPLN waveguide,” Electron. Lett. 39, 1144–1145 (2003).
2. X. Wang, L. Huang, K. Yi, X. Feng, and S. Gao, “All-optical wavelength conversion and five-channel multicasting for 20 Gbit/s QPSK signals in a silicon waveguide,” Opt. Lett. 39(21), 6122–6125 (2014). [PubMed]
3. X. Wang, L. Huang, and S. Gao, “Low-power-penalty wavelength multicasting for 36 Gbit/s 16-QAM coherent optical signals in a silicon waveguide,” Opt. Lett. 39(24), 6907–6910 (2014). [PubMed]
4. P. J. Winzer and R. J. Essiambre, “Advanced modulation formats for high-capacity optical transport networks,” J. Lightwave Technol. 24, 4711–4728 (2006).
5. M. Pu, H. Hu, H. Ji, M. Galili, L. K. Oxenløwe, P. Jeppesen, J. M. Hvam, and K. Yvind, “One-to-six WDM multicasting of DPSK signals based on dual-pump four-wave mixing in a silicon waveguide,” Opt. Express 19(24), 24448–24453 (2011). [PubMed]
6. H. Ji, M. Galili, H. Hu, M. Pu, L. K. Oxenlowe, K. Yvind, J. M. Hvam, and P. Jeppesen, “1.28-Tb/s demultiplexing of an OTDM DPSK data signal using a silicon waveguide,” IEEE Photonics Technol. Lett. 22, 1762–1764 (2010).
7. J. A. Bingham, “Multicarrier modulation for data transmission: An idea whose time has come,” IEEE Commun. Mag. 28, 5–14 (1990).
8. W. Shieh, X. Yi, Y. Ma, and Q. Yang, “Coherent optical OFDM: has its time come?” J. Opt. Netw. 7, 234–255 (2008).
9. J. Armstrong, “OFDM for optical communications,” J. Lightwave Technol. 27, 189–204 (2009).
10. Y. Ma, Q. Yang, Y. Tang, S. Chen, and W. Shieh, “1-Tb/s single-channel coherent optical OFDM transmission with orthogonal-band multiplexing and subwavelength bandwidth access,” J. Lightwave Technol. 28, 308–315 (2010).
11. F. Horlin, J. Fickers, P. Emplit, A. Bourdoux, and J. Louveaux, “Dual-polarization OFDM-OQAM for communications over optical fibers with coherent detection,” Opt. Express 21(5), 6409–6421 (2013). [PubMed]
12. A. J. Lowery, L. B. Du, and J. Armstrong, “Performance of optical OFDM in ultralong-haul WDM lightwave systems,” J. Lightwave Technol. 25, 131–138 (2007).
13. D. Qian, M. F. Huang, E. Ip, Y. K. Huang, Y. Shao, J. Hu, and T. Wang, “High capacity/spectral efficiency 101.7-Tb/s WDM transmission using PDM-128QAM-OFDM over 165-km SSMF within C-and L-bands,” J. Lightwave Technol. 30, 1540–1548 (2012).
14. E. Giacoumidis, A. Tsokanos, C. Mouchos, G. Zardas, C. Alves, J. L. Wei, J. M. Tang, C. Gosset, Y. Jaouën, and I. Tomkos, “Extensive comparison of optical fast OFDM and conventional OFDM for local and access networks,” IEEE J. Opt. Commun. Netw. 4, 724–733 (2012).
15. C. Li, C. Gui, X. Xiao, Q. Yang, S. Yu, and J. Wang, “On-chip all-optical wavelength conversion of multicarrier, multilevel modulation (OFDM m-QAM) signals using a silicon waveguide,” Opt. Lett. 39(15), 4583–4586 (2014). [PubMed]
16. A. Czylwik, “Comparison between adaptive OFDM and single carrier modulation with frequency domain equalization,” IEEE Veh. Technol. Conf.2, 865–869 (1997).
17. E. Giacoumidis, J. L. Wei, X. L. Yang, A. Tsokanos, and J. M. Tang, “Adaptive-modulation-enabled WDM impairment reduction in multichannel optical OFDM transmission systems for next-generation PONs,” IEEE Photonics J. 2, 130–140 (2010).
18. F. Peng, J. Zhang, and W. E. Ryan, “Adaptive modulation and coding for IEEE 802.11 n,” in IEEE Wirel. Commun. Conf. (2007), 656–661.
19. J. M. Tang, P. M. Lane, and K. A. Shore, “Transmission performance of adaptively modulated optical OFDM signals in multimode fiber links,” IEEE Photonics Technol. Lett. 18, 205–207 (2006).
20. E. Giacoumidis, J. L. Wei, X. L. Yang, A. Tsokanos, and J. M. Tang, “Adaptive-modulation-enabled WDM impairment reduction in multichannel optical OFDM transmission systems for next-generation PONs,” IEEE Photonics J. 2, 130–140 (2010).
21. S. Gao, E.-K. Tien, Y. Huang, and S. He, “Experimental demonstration of bandwidth enhancement based on two-pump wavelength conversion in a silicon waveguide,” Opt. Express 18(26), 27885–27890 (2010). [PubMed]
