We propose a novel guard-band-shared direct-detection (GBS-DD) scheme to improve the receiver spectrum efficiency (SE). The 100-Gb/s signal is modulated by 2 sub-bands, which are assigned onto two orthogonal polarizations. The central wavelengths of the two sub-bands are set as 10.84-GHz frequency space. The two sub-bands are then received simultaneously using a single conventional photodiode (PD) of 40-GHz bandwidth. Only one optical pilot carrier is inserted to beat with the 2 sub-bands on the two polarizations. When the 2 sub-band signal entering into the receiver, the signal-to-signal beat interference (SSBI) terms fall and overlap in the same guard band. As a consequence, the bandwidth usage of the PD is enhanced from 1/2 to 2/3. The 100-Gb/s signal is modulated using orthogonal frequency-division multiplexing based on offset quadrature-amplitude-modulation of 64-quadrature amplitude modulation (OFDM/OQAM-64QAM), and transmitted over 80-km standard single mode fiber (SSMF) within a 50-GHz optical grid. It is shown that the proposed GBS-DD scheme can be implemented by the current commercial optical/electrical devices.
© 2014 Optical Society of America
With the emergence of internet social networking sites, mobile phones with internet access and expansion of voice and video communication service, the internet has become an indispensable part of people’s daily life. The excessive growth of Internet traffic is pushing optical communication systems towards higher speed and higher capacity per wavelength channel. In recent years due to its high spectral efficiency (SE), dispersion resilience, and flexibility in frequency multiplexing, coherent optical orthogonal frequency division multiplexing (CO-OFDM) has witnessed a feasible solution to realize multi-Terabit capacity and thousands of kilometers reach transmission [1–3]. A number of sophisticated optical/electrical devices and components are involved, such as four digital-to-analog converters (DACs), a dual-polarization I/Q modulator, four radio-frequency (RF) amplifiers and several Erbium Doped Fiber Amplifiers (EDFAs) at the transmitter side; two polarization beam splitters (PBSs), a local oscillator (LO) laser, two 24 90° optical hybrids, four balanced photodiodes (PDs) with four trans-impedance amplifiers and four analog-to-digital converters (ADCs) at the receiver side.
Recently, cost-effective short and medium reach networks such as data centre, access and metro need to increase the data rate per wavelength to 40-Gb/s or even 100-Gb/s to meet the ever-increasing data traffic demand [4–8]. Direct detection (DD) systems could provide a more cost-effective solution than coherent detention systems for such networks due to the simple system configuration and the lower the number of optical and electrical components required. By employing the traditional intensity modulation/ direct detection (IM/DD) system, W. Yan et al. have demonstrated a 100-Gb/s optical IM/DD transmission with 10G-class devices using 65 GS/s Fujitsu high-speed ADC . However, such IM/DD scheme highly relies on the development of electronics devices/components such as ultra-high-speed ADCs  and the transmission distance is highly limited by the fiber chromatic dispersion. To overcome these two drawbacks, another alternative DD scheme, a host of self-coherent system [5–8] is proposed for short-reach applications since it can significantly lower the expense compared with coherent counterpart while achieving both high data rate and moderate reach. In our previous work, we has also demonstrated that 6 sub-bands 100-Gb/s signal is modulated by one single polarization I/Q modulator, transmitted over 320-km standard single mode fiber (SSMF) and received simultaneously with only one conventional 40-GHz PD . However, the required optical bandwidth is more than 600-GHz.
In this paper, we propose a novel DD scheme for cost-efficient short reach networks, named “guard-band-shared direct detection (GBS-DD)”, to receive 100-Gb/s optical signal using only one conventional 40-GHz PD within a 50-GHz optical grid. Orthogonal frequency division multiplexing/offset quadrature–amplitude-modulation (OFDM/OQAM) is selected as the modulation format to provide the signal spectrum with high side-lobe suppression ratio, which can effectively reduce the electrical sub-band frequency interference in the receiver . Only one optical pilot carrier is inserted to beat with the 2 sub-bands in the two polarizations. When the 2 sub-band signal entering into the receiver, the signal-to-signal beat interference (SSBI) terms fall and overlap in the same guard band. As a consequence, the bandwidth usage of the PD is enhanced from 1/2 to 2/3. The distance of the 100-Gb/s signal transmission over SSMF is up to 80-km.
2. Principle of the novel guard-band-shared DD-OFDM/OQAM system
In the traditional one sub-band pilot-assisted DD optical system, only less than half of the PD bandwidth can be used [10, 12]. Thus, it is practically impossible to transmit and detect 100-Gb/s signal simultaneously by employing the current electrical/optical components. In order to promote the PD bandwidth utilization and overcome the PD bandwidth limitation, a guard-band-shared scheme based on multi-band modulation to carry out 100-Gb/s signal is proposed and demonstrated . In , the signal is first split into 6 sub-bands with equal electrical bandwidth. Each sub-band is assigned with an individual pilot carrier. At the receiver side, the 6 sub-bands are sent into one PD and detected at the same time. However, the optical spectrum efficiency (SE) is quite low, because each optical sub-band has to be > 100-GHz in order to eliminate the cross sub-band beating terms. Figure 1 gives our proposed method that the 2 sub-bands optical signals are allocated in x-/y- polarizations. Only one optical pilot carrier is required to support both x- and y- polarizations. The process of pilot carrier beating with the signal is as follows: one guard band BS is needed. After signal entering into the single PD, two sorts of beating terms are generated, inner sub-band and cross sub-band beating terms. For the inner sub-band beating terms, each sub-band will generate a SSBI, which falls into the same guard band with the frequency from 0 to BS. Because the 2 sub-bands are individually modulated onto two orthogonal polarizations, the sub-band signal only beat with the corresponding carrier when 2 sub-band optical signal entering into the PD. No cross sub-band beating terms will be generated. In such way, only a small guard interval is needed between the 2 sub-bands. Compared with the scheme in , the proposed scheme not only enhances the usage efficiency of PD bandwidth to 2/3, but also can be implemented within a 50-GHz optical grid. Moreover, the proposed scheme can be applied onto wavelength division multiplexing (WDM) transmission. 50-GHz optical multiplexers/de-multiplexers are required to combine/separate the channel signal. Compared to , much more optical channels can be supported in the WDM system using the proposed method.
3. Experimental setup
The experimental setup of the proposed 100-Gb/s GBS-DD OFDM/OQAM system is shown in Fig. 2. At the transmitter, three external-cavity lasers (ECLs) with linewidth of 100 kHz are employed as the light sources. ECL1 (at 1549.87 nm) and ECL2 (at 1549.789nm) are fed into two single polarization I/Q modulators to carry the sub-band signal. Two arbitrary waveform generators (AWGs) running at 12 GS/s sampling rate produce OFDM/OQAM-64QAM RF signal for each modulator. The ECL3 (at 1550 nm) is evenly split into two orthogonal polarizations using a PBS. The upper polarization branch is coupled with the corresponding sub-band signal using a polarization maintaining optical coupler (PMOC), and then combined with the lower branch by a polarization beam combiner (PBC). The combined optical spectrum is shown in Fig. 2(a). A 390-MHz guard band between the 2 sub-bands is reserved to prevent the wavelength variations of the two source lasers. Each sub-band is comprised of 222 subcarriers with OFDM/OQAM-64QAM loading, 4 of which are selected as the pilots to estimate the phase noise. The central subcarrier is unloaded to avoid the direct current (DC) influence. The modulated signal is converted to time domain via an IFFT of size 256. Thus the baseband bandwidth of each sub-band is 223(numbers of subcarriers including the central one)/256(FFT size) × 12(sampling rate) = 10.45-GHz. For the channel estimation, we employ several pairs of training symbols (TSs) in a fashion of [A 0; 0 A], where ’A’ denotes an independent known OFDM/OQAM symbol. In this experiment, 10 TSs are periodically inserted in the front of each OFDM/OQAM frame, which is then followed by 500 payload symbols. The OFDM symbol time duration is 0.46 ps. It is worth noting that, no cyclic prefix is used in OFDM/OQAM modulated signal. The specially designed digital prototype filter in the transmitter/receiver signal processing is designed to combat the inter symbol interference and inter carrier interference , which is a square-root raised cosine filter with roll-off factor of 0.5. The net rate of OFDM/OQAM signal of the 2 sub-bands can be calculated as follows: (222-4)(numbers of information subcarriers)/256 × 12(sampling rate) × 500(payload symbols)/510(symbols of payloads and trainings) × 6(bits per sample of 64-QAM) × 2(dual polarization) × 0.833(20% FEC limit) = 100.18-Gb/s after 20% FEC limit. The transmission link is constructed by a span of 80-km SSMF with EDFA amplification support. A single 40-GHz bandwidth PD is used to detect the entire 2 sub-bands OFDM/OQAM signal. The received electrical spectrum is shown in Fig. 2(b). The signal is sampled by a Tektronix real-time oscillator scope operating at 100-GS/s with 33-GHz electrical bandwidth. The off-line digital signal processing is done with MATLAB program orderly, which includes: 1) carrier frequency offset estimation and OFDM window synchronization; 2) digital filter designed (for OFDM/OQAM only); 3) fast Fourier transform (FFT); 4) channel estimation and phase noise estimation; 5) constellation decision and bit-error-rate (BER) calculation.
4. Results and Discussions
In this paper, OFDM/OQAM-64QAM is selected as the modulation format for the proposed 100-Gb/s signal transmission system. We compare the performance of OFDM/OQAM with conventional OFDM. For conventional OFDM multi-band scheme, a guard band is required to avoid inter symbol interference and crosstalk. Figure 3(a) and 3(b) shows the electrical spectra of conventional OFDM and OFDM/OQAM for 2 sub-bands power loading scheme. Due to the inherent property of high side lobe suppression ratio, the OFDM/OQAM signal provides nearly rectangular spectrum, which is able to effectively reduce the channel crosstalk. We also conduct an experiment to investigate the crosstalk influence. As shown in Fig. 3(c) and 3(d), the signal-to-noise ratio (SNR) for OFDM/OQAM scheme is stabilized at the average level of ~18.45/18.68 dB for all the measured subcarriers of the 2 sub-bands, which suggests that negligible interference from the other band is introduced. Then we change the modulation format to conventional OFDM. While for conventional OFDM the average SNR is ~17.28 dB for the 1st band and ~17.46 dB for the 2nd band. The worst SNRs for the 1st band of OFDM/OQAM and conventional OFDM are 16.13 dB and 12.34 dB, respectively. Meanwhile, 15.92 dB and 13.39 dB are observed for the 2nd band. Obvious crosstalk is observed for conventional OFDM scheme (seen in Fig. 3(c) and 3(d)) at the edges of the 2 sub-bands. It is proved that compared with conventional OFDM, a much smaller guard band is needed for OFDM/OQAM modulated multi-band scheme. Our previous work also has shown that OFDM/OQAM has superiority to load high order QAM format for multi-band power loading transmission system without distortions induced by the electrical low pass filters at the transmitter .
We measure the BER performance as a function of the received power sensitivity for 100 Gb/s OFDM/OQAM and conventional OFDM at back-to-back, as shown in Fig. 4. Under the received power of 8.2 dB, the BER of OFDM/OQAM is 1.84 × 10−2, below 20% FEC limit (BER = 2 × 10−2), while the measured BER of conventional OFDM cannot achieve the 20% FEC limit, as shown in Fig. 4. Compared with conventional OFDM, about 1 dB power improvement is observed for OFDM/OQAM at the BER level of 3 × 10−2. The price paid for OFDM/OQAM is the induced computational complexity.
Optical carrier to signal power ratio (CSPR) is one of the most important issues in optical DD system . We first investigate the BER performance versus CSPR for only one sub-band power loading at back-to-back. As shown in Fig. 5, the individual sub-band 1 and 2 has almost the same optimized CSPR of 10.8 dB. The corresponding BER performance is 1.1 × 10−2. Then, we test the BER performance of the entire 2 sub-bands. Compared to the single band detection, the optimized averaged CSPR for all the 2 sub-bands is slightly reduction to 9.7 dB due to less signal power in the 2-band detection case. The corresponding averaged BER is 1.84 × 10−2.
We further evaluate the transmission performance of the entire 2 sub-bands over 80-km SSMF link. The signal powers of the 2 sub-bands are maintained as a constant in the transmitter, and the averaged CSPR is adjusted by changing the carrier power. We also tune the variable optical attenuator to control the launch power. Figure 6 shows the averaged BER versus launch power with several CSPR values. Under the optimized CSPR of 8.3 dB, the BER is ~1.96 × 10−2, when the launch power is 9.5 dBm.
Under the optimized averaged CSPR, the BER performances of the both 2 sub-bands are listed in Table 1. At back-to-back, for OFDM/OQAM the BERs of the 2 sub-bands are 1.81 × 10−2 and 1.86 × 10−2, while for conventional OFDM are 2.58 × 10−2 and 2.84 × 10−2. We also transmit the signal through 80-km SSMF. The BERs of both 2 sub-bands are 1.95 × 10−2 and 1.98 × 10−2 respectively, which are still under the 20% FEC threshold (BER = 2 × 10−2). The constellations of the recovered OFDM/OQAM-64QAM signal at back-to-back and after 80-km transmission are also shown in Fig. 7. It is expected to further extend the transmission reach when applying lower order modulation format (such as 32-QAM and 16-QAM) and occupying larger optical/electrical bandwidth.
To evaluate the effect on net data rate and achievable transmission distance with different modulation formats, we conduct another measurement of BER versus transmission distance with 32-/16-QAM formats, as shown in Fig. 8. With 20% FEC loading, the net rate is 83.48 Gb/s for 32-QAM, and 66.78 Gb/s for 16-QAM. The maximum SSMF transmission reaches for the two cases are 240-km and 640-km with EDFA amplification. The measured 32-/16-QAM constellations at back-to-back are also shown in the inserts.
We experimentally demonstrated a 100-Gb/s optical OFDM/OQAM-64QAM signal over 80-km SSMF with a single PD within a 50-GHz optical grid based on a novel GBS-DD scheme by employing the current commercial optical/electrical components. It has been shown that OFDM/OQAM is an efficient modulation format to eliminate the channel crosstalk in multi-band optical transmission system.
This work is supported by the National Basic Research (973) Program of China (2010CB328300), and 863 Program of China (2012AA011302).
References and links
1. W. Shieh and C. Athaudage, “Coherent optical orthogonal frequency division multiplexing,” Electron. Lett. 42(10), 587–589 (2006). [CrossRef]
2. D. Qian, M. Huang, E. Ip, Y. 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(10), 1540–1548 (2012). [CrossRef]
3. S. Zhang, M. F. Huang, F. Yaman, E. Mateo, D. Qian, Y. Zhang, L. Xu, Y. Shao, I. B. Djordjevic, T. Wang, Y. Inada, T. Inoue, T. Ogata and Y. Aoki, “40× 117.6 Gb/s PDM-16QAM OFDM transmission over 10,181 km with soft-decision LDPC coding and nonlinearity compensation.” in OFC'2012, paper PDP5C.4 (2012).
4. 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 GSamples/s CMOS DAC Core.” in OFC'2013, paper OM3H.1 (2013).
5. B. Schmidt, Z. Zan, L. B. Du, and A. J. Lowery, “120 Gbit/s over 500-km using single-band polarization-multiplexed self-coherent optical OFDM,” J. Lightwave Technol. 28(4), 328–335 (2010). [CrossRef]
6. X. Chen, A. Li, D. Che, Q. Hu, Y. Wang, J. He, and W. Shieh, “High-speed Fading-free Direct Detection for Double Sideband OFDM Signal via Block-wise Phase Switching.” in OFC'2013, PDP5B.7 (2013)
8. D. Che, A. Li, X. Chen, Q. Hu, Y. Wang, and W. Shieh, “160-Gb/s Stokes Vector Direct Detection for Short Reach Optical Communication.” in OFC'2014, PDP Th5C.7 (2014).
9. 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.” ISSCC, No. EPFL-CONF-190728 (2013).
10. X. Zhang, Z. Li, C. Li, M. Luo, H. Li, C. Li, Q. Yang, and S. Yu, “Transmission of 100-Gb/s DDO-OFDM/OQAM over 320-km SSMF with a single photodiode,” Opt. Express 22(10), 12079–12086 (2014). [CrossRef] [PubMed]
11. Z. Li, T. Jiang, H. Li, and X. Zhang, “Experimental demonstration of 110-Gb/s unsynchronized band-multiplexed superchannel coherent optical OFDM/OQAM system,” Opt. Express 21(19), 21924–21931 (2013). [CrossRef] [PubMed]
12. 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]
13. S. Pierre, C. Siclet, and N. Lacaille, “Analysis and design of OFDM/OQAM systems based on filterbank theory,” IEEE Trans. Signal Process. 50(5), 1170–1183 (2002). [CrossRef]
14. C. Li, X. Zhang, H. Li, C. Li, M. Luo, Z. Li, J. Xu, Q. Yang, and S. Yu, “Experimental Demonstration of 429.96-Gb/s OFDM /OQAM-64QAM over 400-km SSMF Transmission within a 50-GHz Grid .” in the processing of IEEE Photon. J. for publication (2014). [CrossRef]
15. W. R. Peng, I. Morita, H. Takahashi, and T. Tsuritani, “Transmission of High-Speed (>100Gb/s) Direct-Detection Optical OFDM Superchannel,” J. Lightwave Technol. 30(12), 733–8724 (2012). [CrossRef]