Coherent optical OFDM (CO-OFDM) has emerged as an attractive modulation format for the forthcoming 100 Gb/s Ethernet. However, even the spectral-efficient implementation of CO-OFDM requires digital-to-analog converters (DAC) and analog-to-digital converters (ADC) to operate at the bandwidth which may not be available today or may not be cost-effective. In order to resolve the electronic bandwidth bottleneck associated with DAC/ADC devices, we propose and elucidate the principle of orthogonal-band-multiplexed OFDM (OBM-OFDM) to subdivide the entire OFDM spectrum into multiple orthogonal bands. With this scheme, the DAC/ADCs do not need to operate at extremely high sampling rate. The corresponding mapping to the mixed-signal integrated circuit (IC) design is also revealed. Additionally, we show the proof-of-concept transmission experiment through optical realization of OBM-OFDM. To the best of our knowledge, we present the first experimental demonstration of 107 Gb/s QPSK-encoded CO-OFDM signal transmission over 1000 km standardsingle-mode-fiber (SSMF) without optical dispersion compensation and without Raman amplification. The demonstrated system employs 2×2 MIMO-OFDM signal processing and achieves high electrical spectral efficiency with direct-conversion at both transmitter and receiver.
© 2008 Optical Society of America
Orthogonal frequency-division multiplexing (OFDM) has emerged to be the leading modulation technology for the wireless and wireline systems in RF domain, and has been incorporated into many communications standards such as IEEE 802.11 a/g. OFDM transmits data through many parallel orthogonal subcarriers, and provides channel equalization with a relatively simple solution in frequency-domain that would be otherwise quite complex with the conventional time-domain equalization. Recently, there have been intense research interests in applying OFDM to optical communications. Optical OFDM (O-OFDM) has shown extreme robustness to fiber chromatic dispersion [1-5] and polarization mode dispersion (PMD) [6-12]. The O-OFDM has additional advantage of achieving high spectral efficiency using higher-order modulation [13-14] enabling dynamic data rate adaptation. The maximum transmission rate demonstrated so far for coherent optical OFDM (CO-OFDM) is 52.5 Gb/s . Even with bandwidth efficient direct-conversion architecture in transmitter and receiver [4,10], the electrical bandwidth required for 107 Gb/s would still be about 15 GHz. The best commercial DACs/ADCs in silicon integrated circuit (IC) are only run at a bandwidth of 6 GHz , indicating that to realize 100 Gb/s CO-OFDM directly is challenging in a cost-effective manner. To overcome this electrical bandwidth bottleneck associated with DAC/ADC devices, we propose and demonstrate 107 Gb/s CO-OFDM systems using the concept of orthogonal band multiplexing to divide the entire OFDM spectrum into multiple orthogonal bands. These multiple OFDM bands with small or zero frequency guard band can be multiplexed and de-multiplexed without inter-band interference due to inter-band orthogonality. With this scheme, a 107 Gb/s CO-OFDM signal is transmitted through 1000-km (10×100 km) standard-single-mode-fiber (SSMF) using only EDFA (non-Raman amplification) and achieves a Q factor of 11.5 dB, without optical dispersion compensation and without a need for a polarization controller at the receiver. Although transmission at 100 Gb/s and above has been demonstrated at longer distance relying on dispersion compensation module and Raman Amplification (RA) in each span [16-19], our work has achieved the first 1000-km transmission without optical dispersion compensation and without RA beyond 100 Gb/s. The OBM-OFDM has two distinct advantages: (i) with band orthogonality, the spectral efficiency is improved by allowing for zero or small guard band, and (ii) OBM-OFDM offers the flexibility of demodulating two OFDM subbands simultaneously with just one FFT whereas three (I)FFTs would be otherwise needed for the same purpose. It is noted that the two-band subcarrier multiplexed OFDM was used in , but without orthogonality between OFDM bands. Another major difference from  is that we propose a systematic band structure in both transmitter and receiver to resolve ADC/DAC bandwidth bottleneck. Especially upon reception, the OBM-OFDM signal is demonstrated to be partitioned into multiple bands using anti-alias filters such that a relatively low speed 20 GS/s ADC is used to receive 100 Gb/s OFDM, compared with using a 50 GS/s ADC for 50 Gb/s OFDM in . Subcarrier multiplexing for optical communications has also been discussed in [3, 20] with direct-detection and without invoking the orthogonality between the subcarriers or OFDM bands.
We would emphasize that the main motivation of this work is the electronic realization of OBM-OFDM to achieve a CMOS-friendly mixed-signal IC solution for a 100 Gb/s OFDM transceiver. Nevertheless, it is instructive to point out that optical realization of OBM-OFDM serves as an alternative to the other spectral efficient multiplexing schemes including coherent WDM , all-optical OFDM , electro-optically subcarrier-multiplexed OFDM , and orthogonal WDM . In particular,  and  have shown 100 Gb/s OFDM experimental transmission with direct-detection. The difference of our work lies in that the basic processing or multiplexing element for OBM-OFDM is a multi-carrier OFDM signal (or band) whereas for the above-mentioned four schemes is a single-carrier signal. The consequences are that (i) for OBM-OFDM, a cyclic prefix is used to ease the tight bit-level synchronization constraint, (ii) for OBM-OFDM, the efficient (I)FFT is conveniently used for modulation and demodulation, and (iii) the OFDM band spectrum is inherently more tightly-bounded than the single-carrier counterpart, and is readily partitioned with electrical anti-alias filters, and subsequently processed with lower-speed DAC/ADCs. Finally, the proposed OBM-OFDM should not be confused with the multi-band OFDM (MB-OFDM) currently pursued by multiband OFDM alliance (MBOA) for the ultra-wide band (UWB) systems . In MB-OFDM, only one band is transmitted at any point of the time as a means of achieving frequency diversity and multiple access whereas in OBM-OFDM, multiple bands are transmitted simultaneously.
2. Principle of orthogonal-band-multiplexed OFDM (OBM-OFDM)
The principle of the OBM-OFDM is to divide the entire OFDM spectrum into multiple orthogonal OFDM (sub) bands. As shown in Fig. 1, the entire OFDM spectrum comprises N OFDM bands, each with the subcarrier spacing of Δf, and band frequency guard spacing of ΔfG. The subcarrier spacing Δf is identical for each band due to using the same sampling clock within one circuit. The orthogonal condition between the different bands is given by
that is, the guard band is multiple (m times) of subcarrier spacing. This is to guarantee that each OFDM band is an orthogonal extension of another. As such, the orthogonality condition is satisfied not only for the subcarriers inside each band, but it is also satisfied for any two subcarriers from different bands, for instance, fi from band 1 and fj from band 2 are orthogonal to each other (Fig. 1), despite the fact that they originate from different bands. The interesting scenario is that m equals to 1 in (1) such that the OFDM bands can be multiplexed/de-multiplexed even without guard band. We call this method of sub-dividing OFDM spectrum into multiple orthogonal bands ‘orthogonal-band-multiplexed OFDM’ (OBM-OFDM). An identical bandwidth-efficient multiplexing scheme for CO-OFDM has been first proposed in  where it is called cross-channel OFDM (XC-OFDM). We adopt the term of OBM-OFDM to stress the bandwidth reduction through sub-banding of the OFDM spectrum.
Upon reception, each OFDM (sub)band can be de-multiplexed using an anti-alias filter slightly wider than the bands to be detected. Fig. 1 shows two approaches for OBM-OFDM detections. The first approach is to tune the receiver laser to the center of each band, and use an anti-alias filter I that low-pass only one-band RF signal, such that each band is detected separately. The second-approach is to tune the receive laser to the center of the guard band, and use an anti-alias filter II that low-pass two-band RF signal such that two bands are detected simultaneously. In either case, the inter-band interference is avoided because of the orthogonality between the neighboring bands, despite the ‘leakage’ of the subcarriers from neighboring bands. By using OBM-OFDM, CO-OFDM at 107 Gb/s can be realized without forcing the DAC/ADC devices to operate at the extremely high sampling rate.
Figures 2(a)-(c) show the conceptual diagrams for implementing the OBM-OFDM using mixed-signal circuit. In Fig. 2(a), each OFDM baseband transmitter is implemented using digital IC design. The subsequent up-conversion, band-filtering and RF amplification can be implemented in RF IC design. The output of the OFDM baseband transmitter will be filtered through an anti-alias filter and up-convert to appropriate RF band with the center frequency from f1 to fN using an IQ modulator or a complex multiplexer, the structure of which is shown in Fig. 2(c). The range of f1 to fN is centered around zero, given by
where fl is the center frequency of the lth OFDM band, Δfb is the band spacing, L is the maximum of the band number. The output of each IQ modulator is a complex value that has real and imaginary parts as shown in Fig. 2(c). These complex signals are further summed up at the output, namely, real and imaginary parts are added up in separate parallel paths. The combined complex OFDM signal will be used to drive an optical IQ modulator to be up-converted to optical domain [25-26]. We note that the negative and positive bands differ only in the sign of quadrature oscillator ‘sin(2πft)’, and subsequently can be combined and implemented with one complex multiplexer by the same up conversion frequency. However, the baseband input ports need simple modifications to include the two bands that are of mirror-image with each other. At the receive end (Fig. 2(b)), the incoming signal is split into multiple sub bands and down-converted to baseband using IQ demodulators. Anti-alias filters should be used to remove unwanted high frequency components at the output of the demodulators. Again similar to the transmitter, the negative and positive bands can be either down-converted separately using a separate complex mixer, or using the same mixer which separates positive and negative bands. It follows that the DAC/ADC only needs to operate at the bandwidth of each OFDM band, which is approximately scaled down by a factor equal to the number of sub bands from the original complete OFDM spectrum. For instance, if the number of sub bands is five, each OFDM band will only need to cover about 7 GHz optical bandwidth for 107 Gb/s data rate with QPSK modulation and polarization multiplexing. The electrical bandwidth required is 3.5 GHz, or half of the OFDM band spectrum by using direct-conversion at transmit and receive. The ADC/DAC with bandwidth of 3.5 GHz can be implemented in today’s technology  and using a wider bandwidth for each OFDM band will reduce the number of the OFDM bands further down to two or three. Subsequently, the architecture shown in Figs. 2(a) and 2(b) are feasible for implementation in mixed-signal CMOS ICs. It is also noted that the number of transmitter bands and receiver bands do not need to be same, as illustrated in Fig. 2 in which two receiver band partitions are shown reflected by two different anti-alias filters used.
3. Experimental setup and description
The OBM-OFDM could be realized using either subcarrier multiplexing  or wavelength multiplexing to patch multiple orthogonal bands into a complete OFDM spectrum (Fig. 1). The transmission performance such as OSNR sensitivity, nonlinearity, and phase noise impact are independent of the means of OBM-OFDM. Although the future of the 107 Gb/s OBM-OFDM implementation will be in the form of electronic mixed signal IC as shown in Fig. 2. We choose optical multiplexing to obtain OBM-OFDM for proof-of-concept demonstration without affecting our investigation of 107 Gb/s CO-OFDM transmission performance. Fig. 3 shows the experimental setup for the 107 Gb/s CO-OFDM transmission. The multi-frequency (five tones to be exact) optical source spaced at 7.5 GHz is generated using cascaded intensity modulator and phase modulator architecture . The tone spacing and arbitrary waveform generator (AWG) sampling clock are locked by a frequency standard of 10 MHz from the synthesizer. The OFDM signal in each individual band is generated by using a Tektronix Arbitrary Waveform Generator (AWG). The time domain OFDM waveform is first generated with a Matlab program with the parameters as follows: total number of subcarriers is 128 with QPSK encoding, guard interval is 1/8 of the observation period, middle 87 subcarriers out of 128 are filled, from which 10 pilot subcarriers are used for phase estimation. The real and imaginary parts of the OFDM waveforms are uploaded into the AWG operated at 10 GS/s to generate two analog signals, which are then fed into I and Q ports of an optical I/Q modulator, to impress the baseband OFDM signal onto five optical tones. The optical output of the I/Q modulator consists of five-band OBM-OFDM signals, each band carrying 10.7 Gb/s. Although the five bands are filled with the same data, this will not affect the performance of the system studied and our subsequent conclusion. Unlike conventional link design, no dispersion compensation module is used in our transmission experiment, leading to fast phase walk-off and de-correlation between neighboring bands. The system performance we investigate here should be a good representation of the scenario for which each band is filled with independent data. This is verified through simulation that shows the Q penalty difference should be less than 0.4 dB.
The optical OFDM signal from the I/Q modulator is then split into two branches that are delay-mismatched by one OFDM symbol period (14.4 ns), and then combined. This is to emulate the polarization diversity transmitter with data rate of 21.4 Gb/s per band. The two polarization components are completely independent due to the delay of 14.4 ns for each OFDM symbol. The signal is further input into a recirculation loop comprising 100-km fiber and an EDFA to compensate the loss. The signal is coupled out from the loop and received with a polarization diversity coherent receiver [7,11] comprising a receive laser, a polarization beam splitter, two hybrids and four balanced receivers. The receive laser is tuned to the center of each band, and the RF signals from the four balanced detectors first pass through the ‘antialias filters I’ with a low pass bandwidth of 3.8 GHz, such that each band is measured independently (this is the first approach with 21.4 Gb/s per detection described in Section 2). The RF signals are then input into a Tektronix Time Domain-sampling Scope (TDS), acquired at 20 GS/s, and processed with a Matlab program using 2×2 MIMO-OFDM models. The 2×2 MIMO-OFDM signal processing involves [10-11] (1) FFT window synchronization using Schmidl format to identify the start of the OFDM symbol, (2) software estimation and compensation of the frequency offset, (3) channel estimation in terms of Jones Matrix H, (4) phase estimation for each OFDM symbol, and (5) constellation construction for each carrier and BER computation. The channel matrix H is estimated by sending 30 OFDM symbols using alternative polarization launch. The total number of OFDM symbols evaluated is 1000. The measurements of low BER in the order of 10-5 are run multiple times. In practice, the training sequence for channel estimation is only used in the acquisition phase, and will not be repeated in the subsequent OFDM blocks and thus is not counted as an overhead. After completion of acquisition, the channel estimation can be performed through pilot subcarriers or decision-feedback.
4. Experimental results and discussion
Figure 4(a) shows the optical spectrum after 1000-km transmission measured with the polarization diversity coherent receiver shown in Fig. 3. It can be seen that five OFDM bands spaced at 7.5 GHz with guard band about 625 MHz (m=8). The entire OFDM spectrum occupies about 37 GHz and rolls off rapidly at the edge. The out-band components are due to the multi-frequency source generation not tightly bounded at 5 tones. This artifact will not exist in the real application using either subcarrier multiplexing or optical multiplexing OBM-OFDM. Fig. 4(b) shows the ‘zoom-out’ optical spectrum using an optical spectrum analyzer. The m of 8 is chosen for convenience. We have conducted a detailed experiment on the system performance as a function of m, which shows the validity of orthogonal condition (Eq.1). The result will be made known in a separate submission.
Figure 5 shows the detected electrical spectrum after using a 3.8 GHz electrical anti-alias filter. This is equivalent to placing a 7.6 GHz optical band-pass filter centered around each OFDM band. The anti-alias filter is critical for OBM-OFDM implementation. As is shown in Fig. 4(a), without electrical anti-alias filter, the electrical spectrum will be as broad as 15 GHz (which is the photodetector bandwidth). Such a broach spectrum will have alias effect if sampled at 20 GS/s, indicating that at least 30 GS/s ADC has to be used. However, the filtered spectrum in Fig. 5 can be easily sampled with 20 GS/s, or even at a lower speed of 10 GS/s. Additionally, despite the fact that there are some spurious components from neighboring band that is leaked at the edge of the 3.5 GHz filter, since they are orthogonal subcarriers to the interested OFDM subcarriers at the center, they do not contribute to the interference degradation. Some unexpected discrete tones are also shown outside of the pass band, which may be due to the sub-harmonics of clock frequency inside the TDS. Nevertheless, they are too weak to cause any detrimental effects.
Tables 1(a) and 1(b) show the performance of five bands at both back-to-back and 1000-km transmission. It can be seen that both polarizations in each band can be recovered successfully, and this is done without a need for a polarization controller at receive. At the reach of 1000 km, all the sub-bands BER are better than 10-3. The difference of BER in each entry is attributed to the tone power imbalance and instability as well as the receiver imbalance for two polarizations.
|BER (x polarization)||4.1×10-4||4.5×10-4||9.6×10-5||1.4×10-5||7.8×10-4|
|BER (y polarization)||6.8×10-5||1.2×10-4||4.2×10-4||4.2×10-4||7.1×10-4|
Figure 6 shows the BER sensitivity performance for the entire 107 Gb/s CO-OFDM signal at the back-to-back and 1000-km transmission with the launch power of -1 dBm. The BER is counted across all five bands and two polarizations. The inset shows the clear constellation at 1000 km with an OSNR of 20.2 dB. The OSNR required for a BER of 10-3 is respectively 17.0 dB and 19.2 dB for back-to-back and 1000-km transmission. Fig. 7 shows the system Q performance of the 107 Gb/s CO-OFDM signal as a function of reach up to 1000 km. The optimal launch power for all reaches is around -1 dBm. The Q above 12.5 dB is estimated with an electrical SNR corresponding to the subcarrier symbol spread in the constellation diagram (Eq. 8 in ). It can be seen that the Q decreases from 16 dB to 11.5 dB when the reach increases from back-to-back to 1000 km. The Q disparity between two polarizations is attributed to the polarization diversity detector imbalance. We note that this is the first 107 Gb/s transmission over 1000 km SSMF fiber without using optical dispersion compensation module and without Raman amplification, in either single-carrier or multi-carrier format, to the best of our knowledge.
We have proposed and elucidated the principle of orthogonal-band-multiplexed OFDM (OBM-OFDM) to subdivide the entire OFDM spectrum into multiple orthogonal bands. As a result, the DAC/ADCs do not need to operate at extremely high sampling rate. The corresponding mapping to the mixed-signal integrated circuit (IC) design is also revealed. Additionally, we show the proof-of-concept transmission experiment through optical realization of OBM-OFDM. To the best of our knowledge, we present the first experimental demonstration of 107 Gb/s CO-OFDM signal transmission over 1000 km standard-single-mode-fiber (SSMF) without optical dispersion compensation and without Raman amplification. The demonstrated system employs 2×2 MIMO-OFDM signal processing and achieves high electrical spectral efficiency with direct-conversion at both transmitter and receiver.
This work was supported by the Australian Research Council (ARC).
References and links
1. W. Shieh and C. Athaudage, “Coherent optical orthogonal frequency division multiplexing,” Electron. Lett. , 42, 587–589 (2006). [CrossRef]
2. A. J. Lowery, L. Du, and J. Armstrong, “Orthogonal frequency division multiplexing for adaptive dispersion compensation in long haul WDM systems,” in Optical Fiber Commun. Conf., Paper PDP39, Anaheim, CA (2006).
4. W. Shieh, X. Yi, and Y. Tang, “Transmission experiment of multi-gigabit coherent optical OFDM systems over 1000 km SSMF fiber,” Electron. Lett. , 43, 183–185 (2007). [CrossRef]
5. S. L. Jansen, I. Morita, N. Takeda, and H. Tanaka; “20-Gb/s OFDM transmission over 4,160-km SSMF enabled by RF-Pilot tone phase noise compensation,” in Optical Fiber Comm. Conf., Paper PDP15, Anaheim, CA (2007).
6. W. Shieh, W. Chen, and R. S. Tucker, “Polarization mode dispersion mitigation in coherent optical orthogonal frequency division multiplexed systems” Electron. Lett , 42, 996–997 (2006). [CrossRef]
7. W. Shieh, X. Yi, Y. Ma, and Y. Tang, “Theoretical and experimental study on PMD-supported transmission using polarization diversity in coherent optical OFDM systems,” Opt. Express 15, 9936–9947 (2007). [CrossRef] [PubMed]
8. N. Cvijetic, L. Xu, and T. Wang, “Adaptive PMD Compensation using OFDM in Long-Haul 10Gb/s DWDM Systems,” in Optical Fiber Comm. Conf., Paper OTuA5, Anaheim, CA (2007).
10. W. Shieh, “Coherent optical MIMO-OFDM for optical fibre communication systems,” workshop 5, European Conference on Optical Communication, Berlin, Germany (2007).
11. S. L. Jansen, I. Morita, and H. Tanaka, “16×52.5-Gb/s, 50-GHz spaced, POLMUX-CO-OFDM transmission over 4,160 km of SSMF enabled by MIMO processing KDDI R&D Laboratories,” in European Conference on Optical Communications, Paper PD1.3, Berlin, Germany (2007).
12. M. Mayrock and H. Haunstein, “PMD Tolerant Direct-Detection Optical OFDM System,” in European Conference on Optical Communication, Paper 5.2.5., Berlin, Germany (2007).
13. X. Yi, W. Shieh, and Y. Ma, “Phase Noise on Coherent Optical OFDM Systems with 16-QAM and 64- QAM beyond 10 Gb/s,” in European Conference on Optical Communication, paper 5.2.3, Berlin, Germany (2007).
14. F. Buchali and R. Dischler, “Optimized sensitivity direct detection O-OFDM with multi level subcarrier modulation,” in Optical Fiber Communication Conf., Paper OMU5, San Diego, CA (2008).
16. P. J. Winzer, G. Raybon, and M. Duelk, “107-Gb/s optical ETDM transmitter for 100 G Ethernet transport,” in the European Conference on Optical Communication, Paper Th4.1.1, Glasgow, Scotland (2005). [CrossRef]
17. C. R. S. Fludger, T. Duthel, D. van den Borne, C. Schulien, E-D. Schmidt, T. Wuth, E. de Man, G. D. Khoe, and H. de Waardt, “10×111 Gbit/s, 50 GHz Spaced, POLMUX-RZ-DQPSK Transmission over 2375 km Employing Coherent Equalisation,” in Optical Fiber Commun. Conf., Paper PDP22, Anaheim, CA (2007).
18. C. Sethumadhavan, X. Liu, E. Burrows, and L. Buhl, “Hybrid 107-Gb/s Polarization-Multiplexed DQPSK and 42.7-Gb/s DQPSK Transmission at 1.4- bits/s/Hz Spectral Efficiency over 1280 km of SSMF and 4 Bandwidth-Managed ROADMs,” in European Conference on Optical Communication, Paper PD 1.9, Berlin, Germany (2007).
19. A. Sano, E. Yoshida, H. Masuda, T. Kobayashi, E. Yamada, Y. Miyamoto, F. Inuzuka, Y. Hibino, Y. Takatori, K. Hagimoto, T. Yamada, and Y. Sakamaki, “30 x 100-Gb/s all-optical OFDM transmission over 1300 km SMF with 10 ROADM nodes,” European Conference on Optical Communication, Paper PD 1.7, Berlin, Germany, 2007.
20. R. Hui, B. Zhu, R. Huang, C. T. Allen, K. R. Demarest, and D. Richards, “Subcarrier multiplexing for high-speed optical transmission,” IEEE/OSA J. Lightwave Technology , 20, 417–427 (2002). [CrossRef]
21. A. D. Ellis and F. C. G. Gunning, “Spectral density enhancement using coherent WDM,” IEEE Photon. Technol. Lett. , 17, 504–506 (2005). [CrossRef]
22. T. Kobayashi, A. Sano, E. Yamada, Y. Miyamoto, H. Takara, and A. Takada, “Electro-optically subcarrier multiplexed 110 Gb/s OFDM signal transmission over 80 km SMF without dispersion compensation” Electron. Lett. , 44, 225–226 (2008). [CrossRef]
23. G. Goldfarb, G. Li, and M. G. Taylor, “Orthogonal wavelength-division multiplexing using coherent detection,” IEEE Photon. Technol. Lett. , 19, 2015–2017 (2007). [CrossRef]
24. A. Batra, J. Balakrishnan, G. R. Aiello, J. R. Foerster, and A. Dabak, “Design of a multiband OFDM system for realistic UWB channel environments,” IEEE Trans. Microwave Theory and Techniques , 52, 20042123–2138. [CrossRef]
26. Y. Tang, W. Shieh, X. Yi, and R. Evans, “Optimum design for RF-to-optical up-converter in coherent optical OFDM systems,” IEEE Photon. Technol. Lett. , 19, 483–485 (2007). [CrossRef]
27. Y. Wang, Z. Pan, C. Yu, T. Luo, A. Sahin, and A.E. Willner, “A multi-wavelength optical source based on supercontinuum generation using phase and intensity modulation at the line-spacing rate,” in Europe Conference on Optical Communication, Paper Th3.2.4, Rimini, Italy (2003).