In this paper, we propose a novel approach to simultaneously receive multi-band 100-Gb/s direct-detection optical signal with only one polarization and one conventional 40-GHz photodiode. The modulation format of orthogonal frequency-division multiplexing based on offset quadrature amplitude modulation (OFDM/OQAM) is selected to provide signal spectrum with high side-lobe suppression ratio, which can effectively reduce the electrical sub-band frequency interference. The whole 100-Gb/s OFDM/OQAM signal is comprised of 6 sub-bands with 16- and 32-QAM formats loading. Only one guard band is required to accommodate the overlapped 6-band signal-to-signal beat interference (SSBI). The receiver bandwidth is mainly limited by the digital storage oscilloscope (DSO) of 33 GHz. The transmission distance over standard single mode fiber (SSMF) is up to 320 km.
© 2014 Optical Society of America
Coherent detection has achieved great success to provide excellent solution for the high-speed and long-haul transmission in the past few years [1–4]. A typical coherent transmitter/receiver involves a number of sophisticated optical and electrical devices and components, such as a dual-polarization IQ modulator, an optical hybrid, four balanced photodiodes (PDs), four ultra-high-speed electrical analog-to-digital converters (ADCs), ultra-large-capacity digital signal processor, etc. As the lab research and industrial production of coherent 100G/200G have been well carried out, on one hand, the optical research is pursuing higher speed per transponder, which highly relies on the development of electronics devices/components, such as ultra-high-speed ADCs . On the other hand, optical research also has turned to seek for the solutions of a cost-effective short-reach high-speed optical communication. Yan et al. have demonstrated a 100 Gb/s optical intensity modulation/direct detection (IM/DD) transmission with 10G-class devices using 65 GS/s Fujitsu high-speed ADC . However, the transmission distance using conventional IM/DD scheme is highly limited by the fiber chromatic dispersion. Another alternative DD schemes studies the pilot-assisted single sideband . Most recently, Chen et al. have proposed a cost-effective 40-Gb/s direct-detection optical orthogonal frequency division multiplexing (DDO-OFDM) transmission system using block-wise phase switching with single polarization and single PD over 80-km standard single mode fiber (SSMF) .
In this paper, we propose a novel approach to receive multi-band 100-Gb/s direct-detection optical signal with single polarization and only one conventional 40-GHz photodiode. We also employ recent demonstrated OFDM/offset quadrature amplitude modulation (OQAM) as the main modulation format to provide the signal spectrum with every high side-lobe suppression ratio, which can effectively reduce the electrical sub-band frequency interference in the receiver . The whole 100-Gb/s OFDM/OQAM signal is comprised of 6 sub-bands, in which 4 bands are loaded with 32-QAM, and the other 2 bands at higher frequency are loaded with 16-QAM due to the roll off of the frequency response. 6 lasers are used as the pilot carriers to beat the signal. Similar to , only one guard band is required to accommodate the overlapped 6-band signal-to-signal beat interference (SSBI). In our demonstration, the receiver bandwidth is only 33GHz, which is mainly limited by the digital storage oscilloscope (DSO). The distance of the transmission over SSMF is up to 320 km.
2. Principle of the guard-band-shared DDO-OFDM/OQAM system
The concept of the proposed DDO-OFDM/OQAM scheme is depicted in Fig. 1.Figure 1(a) shows a typical pilot-assisted DDO-OFDM/OQAM system. The bandwidth of each optical OFDM/OQAM sub-band is BS. A guard-band with minimum bandwidth of BS is required to accommodate the SSBI . In another word, less than half of the photodiode bandwidth can be used in this scheme. Figure 1(b) shows our proposed DDO-OFDM/OQAM architecture. We first split the signal into multiple (N) sub-bands with equal bandwidth of BS. Each sub-band is assigned with an individual pilot carrier to form a sub-channel. The sub-channels are set far away from each other (≥100 GHz). For the i-th sub-channel, the bandwidth of the guard band set as i × BS.
Two sorts of beating terms are generated after the optical-to-electrical conversion, inner-subchannel and cross-subchannel beating. For the inner-subchannel beating, each subchannel will generate a SSBI, which falls into a same guard-band with the frequency from 0 to BS, while signal bands are allocated consecutively. Because the subchannel spacing is set far away from each other, the cross-subchannel beating terms are outside of the photodiode bandwidth, and then filtered. In this case, only one guard-band with bandwidth of BS is required to be shared for all the SSBI terms. The efficiency of receiver bandwidth usage is greatly enhanced from 1/2 to N/(N + 1). Compared to , the implementation complexity is greatly reduced. Only one photodiode is used to recover the multi-band signal. No optical filter or electrical mixer is required. However, the optical spectral efficiency is much reduced in our proposed method because more optical channels have to be used.
Compared to , the transmission distance using our proposed technique can be greatly extended to several hundreds of kilometers after channel compensation based on digital signal processing (DSP). However, the main drawback of this approach is that more transmitters have to be used. Therefore, the configuration of the transmitter is more complex. Figure 2 gives an efficient way to implement such configuration based on the comb generation. The first comb source I is used to generate the pilot carriers with frequency spacing of fG. Another comb source II with frequency spacing of fS (fS = fG + BS) can be used to carry the sub-band signals in the traditional OFDM superchannel manner . The comb source II is frequency-shifted of BS, and then loaded with OFDM/OQAM signals. Currently, tunable optical frequency comb source can be integrated into a small package device without any radio frequencyinput . Combining with the optical modulator array , our proposed scheme is foreseeable to be integrated into a device with a small package.
The selection of modulation format is critical. Unlike the conventional orthogonal band-multiplexed OFDM system, OFDM/OQAM system has distinguished rectangular spectral shape with side lobe suppression ratio > 35 dB . No timing or frequency synchronization is required to construct the superchannel. When employing such technique in the proposed shared guard-band DDO-OFDM/OQAM system, only very trivial guard-band spacing between different electrical sub-bands is required without introducing any inter-frequency interference.
3. Experimental setup
The experimental setup of the proposed DDO-OFDM/OQAM transmission system is illustrated in Fig. 3.We use 6 bands to carry the 100-Gb/s signal, in which the first 4 bands are loaded with 32-QAM, and the latter 2 bands at higher frequency are loaded with 16-QAM. In the experiment, we use individual lasers as the carriers and pilots instead of the comb lines so that the power of pilot carriers and sub-bands can be easily adjusted. For the first 4 bands, four external-cavity lasers (ECLs) with line-width less than 100 kHz in the upper path are coupled by a 4 × 1 polarization maintaining optical coupler (PMOC), and then fed into an optical IQ modulator. An arbitrary waveform generator running at 10 GS/s is used to produce OFDM/OQAM-32QAM RF signal. Each OFDM/OQAM sub-band is comprised of 114 subcarriers. The FFT size is 256. The middle one subcarrier is unloaded to avoid the DC influence. 2 subcarriers are selected as the pilots to estimate the phase noise. For the channel estimation, we employ several pairs of training symbols (TSs) in a fashion of A, where ‘A’ denotes an independent OFDM/OQAM symbol, which is the same as the one given in . In this experiment, 10 TSs are periodically inserted in the front of each OFDM/OQAM frame, which is then followed by 500 payload symbols. No cyclic prefix is needed in OFDM/OQAM system. The prototype filter is specially designed to combat the inter-symbol interference and inter-carrier interference . The specially designed prototype filter is a square-root raised cosine filter with roll factor of 0.5. The information rate per sub-band (excluding 20% forward error code overhead) in the upper branch is 17.87 Gb/s. Another two lasers are configured in the similar manner to form the lower path. Due to the lower sensitivity response in the higher frequency range of the receiver, 16-QAM is loaded for these two sub-bands. The information rate per sub-band for these two sub-bands is 14.29 Gb/s. Therefore, the total information rate is 100.06 Gb/s. Compared to , We additionally use 6 ECLs as the pilots, which are evenly inserted to support the 6 modulated sub-bands. The optical spectrum at the transmitter is shown in Fig. 3(a). The bandwidth of each sub-band is 4.49 GHz. The pilot carrier spacing is 100 GHz. The bandwidth of the gaps between the pilot and signal carriers for the 6 sub-bands are set as 6.95 GHz, 11.65 GHz, 16.35 GHz, 21.05 GHz, 25.75 GHz and 30.45 GHz, respectively, including the ~200-MHz guard band to prevent the laser frequency variation. To make sure each band is uncorrelated to each other, the 6 sub-bands are first de-multiplexed using a 100-GHz grid optical inter-leaver, and then de-correlated using multiple fiber spans with different length. The transmission link is constructed by 320-km SSMF with Raman amplification. In the receiver, a single PD is used to detect the entire 6 sub-band optical OFDM/OQAM signal, which is then sampled by a digital storage oscilloscope (DSO Tektronix DSA72004B) operating at 100 GS/s. The electronic spectrum at the receiver is shown in Fig. 3(b). Off-line DSP is done in the MATLAB program.
4. Results and discussions
We first generate sub-band signal with even power, and then test the receiver frequency response as a function of frequency by coupling two individual lasers. The frequency is controlled by adjusting the laser wavelengths. The bandwidth of the PD is 40 GHz. However, the bandwidth of DSO is only 33 GHz, which is the main restriction factor limiting the data rate. The amplitude-frequency response is illustrated in Fig. 4.The amplitude response is ~53 dB from 4 to 22 GHz, while gradually decreases ~4 dB from 22 to 32 GHz. Base on such response, we select the first 4 bands to load with 32-QAM, and latter 2 bands to load with 16-QAM. In addition, we pre-emphasize the power of the last two 16-QAM sub-bands at the transmitter to resist the frequency roll off.
Optical carrier to signal power ratio (CSPR) is one of the most important issues in optical direct detection transmission . We first investigate the BER performance versus CSPR in back-to-back. The CSPR of each band is measured individually. Figure 5 shows the BER as a function of CSPR for each sub-band. The first 4 sub-bands with 32-QAM have almost the same optimum CSPR of 14 dB and the best BER performance is 4 × 10−3. The 5th and 6th sub-band with 16-QAM format have the same optimum CSPR of 15.2 dB, which is 1.2 dB higher than the first four sub-bands. It is also shown that the BER performance of the 5th and 6th sub-bands is better than that of the first 4 sub-bands. The BER performance degradation in the last 16-QAM sub-band is mainly due to the high frequency roll off in the DSO.
According to those optimum CSPR conditions, we then set the CSPR of the first 4 sub-bands 1.2 dB lower than the last 2 sub-bands. The BER versus averaged CSPR for the entire 6 bands is shown in Fig. 6.Here, the averaged CSPR is denoted as the ratio of all the pilot carrier power to the entire OFDM/OQAM signal power. The optimum averaged CSPR for all the 6 sub-bands is 14.2 dB. The corresponding BER is 7.1 × 10−3.
We further evaluate the transmission performance of all the sub-bands over 320-km fiber links. We maintain the signal power of all the sub-bands as a constant in the transmitter, and adjust averaged CSPR by changing the carrier power. We also tune the variable optical attenuator to control the launch power. Figure 7 shows the averaged BER versus launch power with several CSPR values. Under the optimum CSPR of 12 dB, the BER is 1.8 × 10−2, when the launch power is 10.2 dBm.
Under the optimal CSPR and launch power condition, the BER performances of all the 6 sub-bands are shown in Fig. 8.The BERs of all the 6 bands are under the 20% FEC threshold (BER = 2 × 10−2) . The constellations of recovered 16- and 32-QAM are also shown in the inserts. Compared to the transmission performance of , two main reasons limit our transmission distance. The first one is that we have loaded higher order modulation format, such as 32-QAM. The second one is that no optical filter is used to extract the carrier and signal before entering into the photodiode. Therefore, more beating ASE noise from multiple channels falls into the receiver, and then degrades the transmission performance.
In this paper, we propose a novel approach to simultaneously receive multi-band 100-Gb/s direct-detection optical signal with only one polarization and one conventional 40-GHz photodiode. The whole 100-Gb/s OFDM/OQAM signal is comprised of 6 sub-bands with 16-/32-QAM format loading. Only one guard-band is required to accommodate the overlapped 6 sub-bands SSBI. The transmission distance over SSMF is up to 320 km. The data rate can be further improved by enhancing the analog bandwidth of the sampling scope and photodiode.
This work is supported by the National Basic Research (973) Program of China (2010CB328300), and 863 Program of China (2012AA011302).
References and links
1. S. J. Savory, G. Gavioli, R. I. Killey, and P. Bayvel, “Electronic compensation of chromatic dispersion using a digital coherent receiver,” Opt. Express 15(5), 2120–2126 (2007). [CrossRef] [PubMed]
2. C. R. S. Fludger, T. Duthel, D. van den Borne, C. Schulien, E.-D. Schmidt, T. Wuth, J. Geyer, E. de Man, G.-D. Khoe, and H. de Waardt, “Coherent equalization and POLMUX-RZ-DQPSK for robust 100-GE transmission,” J. Lightwave Technol. 26(1), 64–72 (2008). [CrossRef]
3. W. Shieh and C. Athaudage, “Coherent optical orthogonal frequency division multiplexing,” Electron. Lett. 42(10), 587–589 (2006). [CrossRef]
4. X. Liu, F. Buchali, and R. W. Tkach, “Improving the nonlinear tolerance of polarization-division-multiplexed CO-OFDM in long-haul fiber transmission,” J. Lightwave Technol. 27(16), 3632–3640 (2009). [CrossRef]
5. L. Kull, T. Toifl, M. Schmatz, P. A. Francese, C. Menolfi, M. Braendli, M. Kossel, T. Morf, T. Meyer Anderson, and Y. Leblebici, “A 90GS/s 8b 667mW 64x interleaved SAR ADC in 32nm digital SOI CMOS.” in Proc. ISSCC (2013), No. EPFL-CONF-190728.
6. W. Yan, T. Tanaka, B. Liu, M. Nishihara, L. Li, T. Takahara, Z. Tao, J. C. Rasmussen, and T. Drenski, “100 Gb/s optical IM-DD transmission with 10G-class devices enabled by 65 G samples/s CMOS DAC Core,” in Proc. OFC (2013), paper OM3H.1. [CrossRef]
7. W. R. Peng, X. X. Wu, V. R. Arbab, K. M. Feng, B. Shamee, L. C. Christen, J. Y. Yang, A. E. Willner, and S. Chi, “Theoretical and experimental investigations of direct-detected RF-tone-assisted optical OFDM systems,” J. Lightwave Technol. 27(10), 1332–1339 (2009). [CrossRef]
8. X. Chen, A. Li, D. Che, Q. Hu, Y. Wang, J. He, and W. Shieh, “Block-wise phase switching for double-sideband direct detected optical OFDM signals,” Opt. Express 21(11), 13436–13441 (2013). [CrossRef] [PubMed]
9. Z. Li, T. Jiang, H. Li, X. Zhang, C. Li, C. Li, R. Hu, M. Luo, X. Zhang, X. Xiao, Q. Yang, and S. Yu, “Experimental demonstration of 110-Gb/s unsynchronized band-multiplexed superchannel coherent optical OFDM/OQAM system,” Opt. Express 21(19), 21924–21931 (2013). [CrossRef] [PubMed]
10. W. R. Peng, H. Takahashi, I. Morita, and H. Tanaka, “Transmission of a 213.7-Gb/s single-polarization direct-detection optical OFDM superchannel over 720-km standard single mode fiber with EDFA-only amplification,” in Proc. ECOC (2010), PDP2.5. [CrossRef]
11. X. Liu, S. Chandrasekhar, P. J. Winzer, T. Lotz, J. Carlson, J. Yang, G. Cheren, and S. Zederbaum, “1.5-Tb/s guard-banded superchannel transmission over 56 100-km (5600-km) ULAF using 30-Gbaud pilot-free OFDM-16QAM signals with 5.75-b/s/Hz net spectral efficiency.” in Proc. ECOC (2012), paper Th.3.C.5.
13. E. Giacoumidis, M. F. O'Keefe, I. Aldaya, R. G. Walker, Y. Zhou, N. Cameron, E. Kehayas, A. Tsokanos, I. Tomkos, N. J. Doran, and L. Zimmermann, “Cost-effective broadband GaAs IQ modulator array for long-reach OFDM-PONs.” in Proc. ECOC (2013), paper We.1.F.4.
14. W. R. Peng, I. Morita, H. Takahashi, and T. Tsuritani, “Transmission of High-Speed (> 100 Gb/s) Direct-Detection Optical OFDM Superchannel,” J. Lightwave Technol. 30(12), 2025–2034 (2012). [CrossRef]
15. E. Torrengo, R. Cigliutti, G. Bosco, G. Gavioli, A. Alaimo, A. Carena, V. Curri, F. Forghieri, S. Piciaccia, M. Belmonte, A. Brinciotti, A. L. Porta, S. Abrate, and P. Poggiolini, “Transoceanic PM-QPSK Terabit superchannel transmission experiments at Baud-rate subcarrier spacing,” in Proc. ECOC (2010), paper We.7.C.2.