The feasibility of implementing 128-QAM in off-the-shelf component-based real-time optical OFDM (OOFDM) transceivers incorporating advanced channel estimation, on-line performance monitoring and live parameter optimisation, is experimentally investigated, for the first time, in intensity-modulation and direct-detection (IMDD) single-mode fibre (SMF) and multi-mode fibre (MMF) transmission systems involving directly modulated DFB lasers. The highest ever spectral efficiency of 5.25bit/s/Hz is demonstrated successfully in the aforementioned simple systems. Experimental investigations show that, it is feasible to transmit 5.25Gb/s 128-QAM-encoded OOFDM real-time signals over 25km MetroCorTM SMFs and 500m 62.5/125μm OM1 MMFs. The impact of key parameters on the transmission performance of the real-time OOFDM transceivers with 128-QAM encoding are explored, based on which optimum signal clipping ratios are identified.
©2009 Optical Society of America
To satisfy the exponentially increasing end-users’ demands for broadband services, optical orthogonal frequency division multiplexing (OOFDM)  has widely been considered as one of the strongest contenders for practical implementation in various cost-sensitive network scenarios such as access networks  and local area networks , this is because OOFDM has unique and inherent advantages, including, for example, potential for providing cost-effective technical solutions by fully exploiting the rapid advances in modern digital signal processing (DSP) technology, and considerable reduction in optical network complexity owing to its great resistance to dispersion impairments and efficient utilization of channel spectral characteristics. Apart from the above-mentioned features, OOFDM is also capable of offering, in both the frequency and time domains, hybrid dynamic allocation of broad bandwidth among various end users . Such ability significantly enhances not only the flexibility of optical networks but also the compatibility with existing optical networks, allowing transparent support for legacy services.
The experimental demonstration of real-time OOFDM transceivers is vital for enabling the realization of the great potential of OOFDM in practical optical networks. Recently, in multi-mode fibre (MMF)-based local area networks [5–7] and single-mode fibre (SMF)-based access networks , we have made significant breakthroughs in experimentally demonstrating, for the first time, real-time OOFDM transceivers based on off-the-shelf components, in which Altera Stratix II GX field-programmable gate arrays (FPGAs) are utilised to perform real-time DSP including most notably the self-developed core logic functions of inverse fast Fourier transform (IFFT) and fast Fourier transform (FFT). In , we proposed a novel pilot subcarrier-assisted channel estimation technique, which offers a number of salient features including high accuracy, low complexity, small pilot bandwidth usage, excellent stability and buffer-free data flow. The channel estimation technique enables not only the use of arbitrary signal modulation formats but also the measurement of the frequency response of an entire transmission system ranging from the IFFT in the transmitter to the FFT in the receiver. It is noted that multi-gigabit real-time OOFDM receivers have been demonstrated experimentally in coherent transmission systems, where off-line DSP approaches are still adopted in corresponding transmitters [9,10].
More recently, significantly improved real-time OOFDM transceivers have also been implemented , which offer advanced functionalities such as on-line performance monitoring and live optimization of key transceiver parameters. On-line monitoring enables the display of individual subcarrier bit error rate (BER), total channel BER and measurement of system frequency response. Conducting live parameter optimisation based on on-line performance monitoring provides an extremely effective means of optimising, during signal transmission, key parameters including, for example, signal clipping, subcarrier amplitude and operating conditions of directly modulated DFB lasers (DMLs). It has been shown  that the improved real-time OOFDM transceiver supports the fastest ever real-time end-to-end transmission of a 6Gb/s 16-quadrature amplitude modulation (QAM)-encoded OOFDM signal over a 300m OM1 MMF with a power penalty of 0.5dB in an intensity-modulation and direct-detection (IMDD) system employing a DML. It should be noted that, a different performance monitoring technique has also been proposed and analyzed in coherent OOFDM transmission systems .
To further improve the transmission capacity of the real-time OOFDM transceivers, one of the most cost-effective approaches is to employ high signal modulation formats such as M-ary QAM. In particular, this approach is extremely valuable for MMF-based transmission systems owing to their lower bandwidths. The use of high signal modulation formats, however, inevitably imposes a number of technical challenges in the transceiver design. The challenges are summarized as followings:
- • Increase in optical signal-to-noise ratio (OSNR). As an example, according to theoretical predictions , for achieving BERs of 1.0×10−3 in M-ary-QAM-encoded OOFDM transmission systems involving digital-to-analogue/analogue-to-digital converters (DACs/ADCs) operating at fixed sampling rates, a 1-bit increase in brings about approximately a 3dB increase in minimum required OSNR. Such an increase may require photo-detectors with better receiver sensitivities to be implemented, as the OOFDM signals have relatively small signal extinction ratios.
- • Increased susceptibility to unwanted DAC/ADC impairments. These unwanted DAC/ADC impairments include sampling jitter, conversion nonlinearity and noise associated with clipping and quantization. In addition, signal distortions and noise introduced by other analogue components such as RF amplifiers have an increasing impact.
- • Increased requirements on accuracy of real-time DSP algorithms, channel estimation and symbol synchronization.
- • Increased requirements on linearity of electrical-to-optical converters such as DMLs.
In previously reported works, the highest signal modulation format used commonly on all the subcarriers within an OOFDM symbol is 64-QAM in an IMDD OOFDM system . Although 128-QAM was also used in a system of such type , the signal modulation format, however, just occupies a small portion of the signal spectral region due to the utilization of adaptive bit-loading algorithms. It should be pointed out, in particular, that all those experimental works [3,14] have been undertaken using off-line DSP approaches, which do not consider the limitations imposed by the precision and speed of practical DSP hardware for realizing real-time end-to-end transmission.
The thrust of the present paper is to explore experimentally, for the first time, the feasibility of implementing 128-QAM in real-time OOFDM transceivers [7,8,11] for transmitting real-time OOFDM signals over simple IMDD SMF/MMF systems involving DMLs. This work also provides an excellent opportunity of rigorously evaluating the self-developed IFFT/FFT logic algorithms and the proposed channel estimation technique. Here it is also worth mentioning that, in comparison with experimental measurements reported in , in this paper, lower DAC/ADC sampling rates of 2GS/s are considered, which give rise to narrower OOFDM signal bandwidths and thus reduces the effect of the DAC-induced system frequency response roll-off . This enables us to highlight the effects associated with high signal modulation formats.
In this paper, it is shown that 128-QAM can be adopted on all the information-carrying subcarriers within a symbol in the real-time OOFDM transceivers without utilizing oversampling  and spectral guard band . A spectral efficiency of 5.25bit/s/Hz is achieved successfully, which is, as far as we are aware, the highest value reported in IMDD OOFDM transmission systems. In DML-based IMDD SMF transmission systems without in-line optical amplification and chromatic dispersion compensation, the real-time 128-QAM OOFDM transceivers support end-to-end transmission of 5.25Gb/s over 25km MetroCorTM SMFs with negative power penalties of −0.5dB at BERs of 1.0×10−3. Whilst in DML-based MMF transmission systems, the transceivers support end-to-end transmission of 5.25Gb/s over 500m MMFs with 0.5dB power penalties at BERs of 1.0×10−3. The impact of key parameters on the transmission performance of the real-time OOFDM transceivers with 128-QAM encoding are also explored, based on which optimum signal clipping ratios are identified.
2. Real-time experimental system setup
Figure 1 shows the real-time experimental system setup, in which use is made of the real-time OOFDM transceiver design similar to that reported in , except that, in this paper, the DAC/ADC sample rates are set to 2GS/s, symbol rate is set to 50MHz and subcarrier peak amplitudes prior to the IFFT in the transmitter are set to be identical. Detailed descriptions have already been made of the real-time transceiver architecture incorporating the IFFT/FFT logic function , the channel estimation technique and its application in measuring the system frequency response , as well as the improved real-time OOFDM transceiver design with on-line performance monitoring and live parameter optimization , therefore, an outline of the experimental system is presented below. In addition, to gain an in-depth understanding of the impact of 128-QAM on the transmission performance, comparisons between 128-QAM and 64-QAM are also presented.
In the real-time OOFDM transmitter, the 32 point IFFT logic function supports 32 equally spaced subcarrier frequencies, of which 15 are located in the positive frequency bins occupying the whole Nyquist band. A 98-bit (84-bit for 64-QAM) wide parallel pseudo random bit sequence of length of 88500 words is employed as information data. One extra parallel bit sequence of a fixed 7-bit (6-bit for 64-QAM) wide pattern is used to represent pilot data. To maximise the SNR of the received pilot data in the receiver, the fixed pilot pattern corresponds to one of the eight (four for 64-QAM) end constellation points with maximum power. Prior to feeding 15 128-QAM (64-QAM) encoders, the pilot data is embedded in the information data in such a way that the pilot data occurs on successive subcarriers in consecutive OOFDM symbols . The 105-bit (90-bit for 64-QAM) wide data constructed from the information and pilot data bit sequences is employed to feed the 15 parallel encoders.
At the input of the IFFT, the aforementioned 15 subcarriers and one extra subcarrier having zero power at zero frequency are arranged to satisfy the Hermitian symmetry with respect to their complex conjugate counterparts. The self-developed IFFT logic function is then employed to perform the IFFT on all the 32 subcarriers. At the output of the IFFT, real-valued OOFDM symbols having 32 signed samples are produced. The amplitude of the output samples is clipped at a fixed but programmable clipping level, this clipped range is then quantized to 8-bits to match the resolution of the employed DAC. The clipping level is stored in the FPGA’s embedded memory and can be updated, during live transmission, to optimise the transceiver performance.
A cyclic prefix of 8 samples is added to each symbol, giving rise to 40 samples per symbol. The internal system clock is set to 50MHz, which is equal to the symbol rate. The 50MHz symbol rate and 40 samples per symbol give a sample rate of 2GS/s. The signed samples are converted to unsigned values by adding an appropriate DC offset. After performing sample ordering and bit arrangement, the unsigned 40 samples are streamed to the DAC interface at 2GS/s. The entire symbol consisting of 320 bits is fed in parallel to 32 high-speed 10:1 dedicated hardware serialisers, the interface thus consists of 4 samples transferred in parallel at a rate of 0.5GHz, giving the required aggregated sample rate of 2GS/s. The DAC generates an analogue electrical OOFDM signal having a maximum peak-to-peak voltage of 636mV. The signal is attenuated as necessary and subsequently, together with an appropriate DC bias current, injected into a 1550nm DML with a 3-dB modulation bandwidth of approximately 10GHz and a maximum optical output power of about 0dBm.
The converted OOFDM signal emerging from the DML is coupled into an erbium-doped fibre amplifier (EDFA) with a 15dB optical gain and a 5dB noise figure. After passing through a 0.8nm optical filter, the amplified optical signal is coupled into MetroCorTM SMFs of different lengths. It should be noted that the use of the EDFA is to vary the optical launch power only. For investigating the performance of the OOFDM signal over a 62.5/125μm OM1 MMF system having a 3-dB optical bandwidth of approximately 675MHz∙km and a linear loss of 0.6dB/km, the MetroCorTM SMF is replaced by the MMF, into which the optical signal is coupled via a mode-conditioning patch cord.
In the receiver, after passing through an optical attenuator, the received OOFDM signal is converted into the electrical domain using a MMF pigtailed 12GHz PIN with TIA. The PIN has a receiver sensitivity of −17dBm (corresponding to 10 Gb/s non-return-to-zero data at a BER of ). The electrical signal is first amplified with a 2.5GHz, 20dB RF amplifier, then attenuated as needed to optimise the signal amplitude to suit the input range of the ADC. After passing through an electrical low-pass filter, the signal is converted via a balun to a differential signal and then digitized by a 2GS/s, 8-bit ADC in the receiver. Finally, the digital samples are fed, via a digital interface identical to the interface of the DAC, to a second Altera Stratix II GX FPGA, which performs, using an inverse process compared to that in the transmitter, the real-time DSP on the received symbols and recovers the data. Pilot subcarrier detection, channel estimation, channel equalization and measurement of system frequency response are performed after the FFT following procedures detailed in [8,11].
Symbol synchronization is performed by continuous transmission of symbols of a fixed pattern over the transmission system. By using the FPGA embedded logic analyser (SignalTap II) with JTAG connection to a PC, the captured samples of the fixed symbols can be viewed, thus the sample offset is determined and subsequently compensated by adjusting the inserted sample offset accordingly. It should be pointed out, in particular, that such a symbol synchronization process is performed only once at the establishment of the transmission connection. In practical transmission systems, unexpected disturbances to the systems may occur, dynamic synchronization is thus required. Very recently we have proposed and theoretically verified a high-speed and accurate dynamic symbol synchronisation technique, which automatically performs clock recovery and symbol alignment in the FPGA of the receiver without requiring highly stable and expensive voltage controlled oscillators. The implementation and transmission performance of such an advanced synchronisation technique in the real-time OOFDM transmission systems is beyond the scope of the present paper and will be reported elsewhere in due course.
The bit error count over 88500 symbols is continuously updated and displayed with the embedded logic analyser for the entire channel and also for each individual subcarrier over the measurement period. This enables fine adjustment of the system parameters to maximize the transmission performance. In addition, the logic analyser also displays and continuously updates the total number of bit errors and the corresponding symbols accumulated since the start of a transmission session. This enables the measurement of the BER of the entire transmission channel at unlimited low values, provided that a sufficiently long operation time is allowed. Clock synthesizers based on a common reference clock are used to generate the system timing for both the transmitter and the receiver.
3. Experimental results
As already discussed in Section 2, with the 50MHz FPGA operating speeds and the 2GS/s sample rates of the DAC/ADC, 5.25Gb/s (4.5Gb/s) OOFDM signals are produced when 128-QAM (64-QAM) is taken on all the 15 information-bearing subcarriers in the positive frequency bins. The 2GS/s sample rate also gives rise to a signal spectral bandwidth of 1GHz, resulting in a spectral efficiency of 5.25bit/s/Hz (4.5bit/s/Hz) for a 128-QAM- (64-QAM)-encoded OOFDM signal. For all the experimental measurements presented in this section, the DAC/ADC resolution is 8 bits, and the DFB laser operates at a bias current of 36mA, under which the optical output power is −3.5dBm. After appropriately adjusting the optical gain of the EDFA followed by a 0.8nm optical filter, the optical power injected into the MetroCorTM SMF is fixed at 7dBm, which is chosen to compensate for approximately 14dB of loss in an entire system involving a 50km MetroCorTM SMF. To ease performance comparisons between SMFs and MMFs without introducing performance degradation, the same optical power is also adopted in the MMF system.
3.1 Signal clipping ratio optimization
It is well known  that the DAC-induced effects of signal clipping and quantization noise are more pronounced for high signal modulation formats. The DAC employed in the transmitter fixes the bits of quantization at 8, which is an optimum value for signal modulation formats up to 256-QAM . Therefore, experimental explorations are undertaken to identify optimum clipping ratios for different signal modulation formats. The definition of signal clipping ratio can be found in . For 128-QAM- and 64-QAM-encoded OOFDM signals, the BER as a function of clipping ratio is shown in Fig. 2 , where the analogue electrical back-to-back case and the 25km MetroCorTM SMF transmission case are considered to distinguish the signal clipping effect induced by the DAC from those induced by optical components. In the first transmission case, the electrical signal from the DAC in the transmitter (point A) is directly connected to the attenuator in the receiver (point B) without any optical components being involved, as shown in Fig. 1.
It can be seen in Fig. 2 that, for 128-QAM and 64-QAM there exist optimum clipping ratios of 11.9dB and 13.5dB, respectively, each of which is identical for the two system configurations. For a specific signal modulation format, when clipping ratios are smaller than the optimum value, the signal is clipped considerably, thus resulting in distorted signal waveforms; whilst when clipping ratios are higher than the optimum value, the signal is quantized at a wide quantization step size over an enlarged dynamic range, thus producing large quantization noise. The co-existence of the effects of signal clipping and quantization noise leads to the occurrence of the optimum clipping ratios, as shown in Fig. 2. Furthermore, for a given signal modulation format, the similarities of the BER evolution curves and in particular, optimum clipping ratios, between the two different configurations considered, indicate that DML nonlinearities do not considerably alter the signal waveform. However, subcarrier × subcarrier intermixing upon direct detection in the PIN causes the occurrence of an unwanted spectral distortion region in the vicinity of the optical carrier frequency [15,16]. As a direct result of the effects of subcarrier intermixing and noise associated with the PIN, for the 128-QAM- (64-QAM-) encoded signal, the minimum BER of 1.6×10−5 (1.1×10−8) observed in the analogue electrical back-to-back case is increased to 9.7×10−4 (3.5×10−5) observed in the 25km MetroCorTM SMF transmission case. In addition, it can also be seen in Fig. 2 that, the optimum clipping ratio decreases for higher signal modulation formats. This is because, when the signal power remains constant, a higher signal modulation format–induced reduction in constellation point spacing makes the signal less tolerant to the quantization noise effect.
Once again, for both of the aforementioned transmission system configurations, constellations of the 128-QAM (64-QAM)-encoded signal clipped at the optimum clipping ratio are recorded after performing channel equalization in the receiver. The superimposed constellations for all the subcarriers are shown in Fig. 3 , in which clear constellations are observed for all these cases. In addition, experimental measurements also show that, for optical input powers of <10dBm, the impact of fibre nonlinearity on the system performance is negligible, implying that the identified optimum clipping ratios are also applicable for the MMF transmission systems.
3.2 Transmission performance over SMFs
Making use of the identified optimum clipping ratio, experimental measurements are performed of the transmission performance of real-time 5.25Gb/s 128-QAM-encoded OOFDM signals in DML-based IMDD 25km MetroCorTM SMF systems without in-line optical amplification and chromatic dispersion compensation. The measured BER as a function of received optical power is shown in Fig. 4(a) , where the BER performance for optical back-to-back is also plotted. In addition, to gain a better understanding of the signal modulation format-dependent transmission performance, Fig. 4(b) presents the BER versus received optical power for transmission of real-time 4.5Gb/s 64-QAM-encoded OOFDM signals over three different systems having configurations similar to those considered in Fig. 4(a). The three systems include optical back-to-back, 25km and 50km MetroCorTM SMFs.
It can be seen in Fig. 4(a) that, for the case of real-time 5.25Gb/s 128-QAM-encoded OOFDM transmission in the 25km MetroCorTM SMF (optical back-to-back) configuration, a BER of 1.0×10−3 (7.8×10−4) is achieved at a received optical power of −7.3dBm (−6.0dBm). On the other hand, as demonstrated in Fig. 4(b), the replacement of 128-QAM with 64-QAM enables the real-time 4.5Gb/s OOFDM transmission over 50km (25km) MetroCorTM SMFs at a BER of 1.9×10−4 (3.2×10−5) at a received optical power of −8.6dBm (−7.1dBm). Furthermore, comparisons of optical back-to-back performances between Fig. 4(a) and Fig. 4(b) indicate that, for achieving a BER of 1.0×10−3, a 3.1dB increase in optical power is required if 64-QAM is replaced by 128-QAM. This is in excellent agreement with theoretical results published previously .
Very similar to those corresponding to real-time 3Gb/s 16-QAM-encoded OOFDM signals , negative power penalties of approximately −0.5dB at 1.0×10−3 are observed in Fig. 4(a) and Fig. 4(b) for all the cases considered. The measured negative power penalty is transmission distance dependent irrespective of signal modulation formats used. This is consistent with our numerical simulation results. The physical mechanism underpinning the occurrence of the negative power penalty is that the negative chromatic dispersion parameter associated with the MetroCorTM SMF can be compensated by the positive transient frequency chirp associated with the DML . Compared to short transmission distances, such compensation can lift up slightly the channel frequency response in the high frequency region for relatively long transmission distances, as seen in Fig. 5(a) . The employed DAC is a major contributor to the channel frequency response roll-off effect .
In Fig. 4(a) and Fig. 4(b), error floors start to develop at BERs of approximately 1.0×10−3. It is worth addressing, in particular, that the occurrence of such trends is not due to the real-time DSP functionalities and digital-to-analogue/analogue-to-digital conversions, as indicated clearly in Fig. 2. The cause of the error floors is the subcarrier intermixing effect mentioned in Section 3.1. This can be understood by considering Fig. 5(b), where the error ratio, which is defined as the percentage ratio between the number of error bits on a given subcarrier and the total number of error bits on all the subcarriers at a BER of 1.0×10−3, is presented to quantify explicitly error distributions for various scenarios. It can be seen in Fig. 5(b) that, almost uniform error ratios occur over the entire subcarriers, even though, according to Fig. 5(a), the low frequency subcarriers experience SNRs roughly 2dB higher than those experienced by the high frequency subcarriers. This implies that the subcarriers in the positive frequency bins suffer right triangle-shaped spectral distortions, i.e., the first subcarrier experiences the strongest spectral distortion effect, which reduces gradually for other subcarriers with higher frequencies. The triangle-shaped spectral distortions offset the channel frequency response-induced SNR variations. In addition, it can also be found from Fig. 4(a) and Fig. 4(b) that, the error floors are extended towards high received optical powers for high modulation formats and long transmission distances. All the above-mentioned characteristics are in line with those associated with subcarrier intermixing. It has already been experimentally verified  that, the subcarrier intermixing effect is negligible for transmission of 3Gb/s 16-QAM-encoded OOFDM signals over MetroCorTM SMFs of up to 75km. Of course, the use of a spectral guard band can reduce such an effect. However, the approach decreases the signal spectral efficiency and increases the transceiver complexity.
3.3 Transmission performance over MMFs
To explore the feasibility of employing 128-QAM in DML-based MMF transmission systems, experimental investigations are also undertaken of real-time 5.25Gb/s 128-QAM-encoded OOFDM transmission over 62.5/125μm OM1 MMF systems. The measured BER versus received optical power is plotted in Fig. 6 for optical back-to-back and two systems of lengths 100m and 500m. It can be seen in Fig. 6 that, for the case of 500m (100m) MMF transmission, a BER of 9.3×10−4 (9.7×10−4) is achieved at a received optical power of −6.0dBm (−6.2dBm). Experiments also show that, in the 500m (100m) MMF system, the use of 64-QAM can reduce the BER to 1.0×10−4 (2.0×10−5) at a received optical power of −4.3dBm (−5dBm). For both transmission systems considered here, similar power penalties of approximately 0.5dB at BERs of 1.0×10−3 are observed, which originate from the co-existence of the effects of differential mode delay (DMD) and modal noise: for a long MMF system, a reduction in the model noise effect  offsets, to some extent, the increased DMD effect.
The feasibility of implementing 128-QAM in off-the-shelf component-based real-time OOFDM transceivers incorporating channel estimation, on-line performance monitoring and live parameter optimisation, has been explored experimentally, for the first time, in simple DML-based IMDD SMF/MMF transmission systems. The highest ever spectral efficiency of 5.25bit/s/Hz has been achieved successfully in an OOFDM transmission system. It has been shown that, real-time transmissions of 5.25Gb/s 128-QAM-encoded OOFDM signals over 25km MetroCorTM SMFs and 500m 62.5/125μm OM1 MMFs are feasible. In addition, detailed discussions have also been made of the impact of key parameters on the BER performance of the real-time OOFDM transceivers. Optimum signal clipping ratios have been identified for various signal modulation formats. This work suggests that it is possible to implement real-time OOFDM transceivers running at 40Gb/s for cost-sensitive access and local area networks, if use is made of multiband transmission and higher DAC/ADC sampling rates.
This work was partly supported by the European Community's Seventh Framework Programme (FP7/2007-2013) within the project ICT ALPHA under grant agreement n° 212 352, in part by the U.K. Engineering and Physics Sciences Research Council under Grant EP/D036976, and in part by The Royal Society Brian mercer Feasibility Award. The work of X.Q. Jin was also supported by the School of Electronic Engineering and the Bangor University.
References and links
1. J. Armstrong, “OFDM for optical communications,” J. Lightwave Technol. 27(3), 189–204 (2009). [CrossRef]
2. T. Duong, N. Genay, P. Chanclou, B. Charbonnier, A. Pizzinat, and R. Brenot, “Experimental demonstration of 10 Gbit/s for upstream transmission by remote modulation of 1 GHz RSOA using adaptively modulated optical OFDM for WDM-PON single fiber architecture,” European Conference on Optical Communication (ECOC), (Brussels, 2008), PD paper Th.3.F.1.
3. S. C. J. Lee, F. Breyer, S. Randel, R. Gaudino, G. Bosco, A. Bluschke, M. Matthews, P. Rietzsch, R. Steglich, H. P. A. van den Boom, and A. M. J. Koonen, “Discrete multitone modulation for maximizing transmission rate in step-index plastic optical fibres,” J. Lightwave Technol. 27(11), 1503–1513 (2009). [CrossRef]
4. D. Qian, J. Hu, P. N. Ji, and T. Wang, “10-Gb/s OFDMA-PON for delivery of heterogeneous services,” Optical Fibre Communication/National Fibre Optic Engineers Conference (OFC/NFOEC), (OSA, 2008), Paper OWH4.
5. R. P. Giddings, X. Q. Jin, H. H. Kee, X. L. Yang, and J. M. Tang, “First experimental demonstration of real-time optical OFDM transceivers”, European Conference on Optical Communication (ECOC) (Vienna, Austria, 2009), Paper 6.6.1.
6. R. P. Giddings, X. Q. Jin, H. H. Kee, X. L. Yang, and J. M. Tang, “Real-time implementation of optical OFDM transmitters and receivers for practical end-to-end optical transmission systems,” Electron. Lett. 45(15), 800–802 (2009). [CrossRef]
8. X. Q. Jin, R. P. Giddings, and J. M. Tang, “Real-time transmission of 3Gb/s 16-QAM encoded optical OFDM signals over 75km SMFs with negative power penalties,” Opt. Express 17(17), 14574–14585 (2009). [CrossRef] [PubMed]
10. S. Chen, Y. Yang, Y. Ma, and W. Shieh, “Real-time multi-gigabit receiver for coherent optical MIMO-OFDM signals,” J. Lightwave Technol. 27(16), 3699–3704 (2009). [CrossRef]
11. R. P. Giddings, X. Q. Jin, and J. M. Tang, “First experimental demonstration of 6Gb/s real-time optical OFDM transceivers incorporating channel estimation and variable power loading,” Opt. Express 17(22), 19727–19738 (2009). [CrossRef] [PubMed]
13. J. M. Tang and K. A. Shore, “Maximizing the transmission performance of adaptively modulated optical OFDM signals in multimode-fiber links by optimizing analog-to-digital converters,” J. Lightwave Technol. 25(3), 787–798 (2007). [CrossRef]
14. C.-T. Lin, S.-P. Dai, W.-J. Jiang, J. Chen, Y.-M. Lin, P. T. Shih, P.-C. Peng, and S. Chi, “Experimental demonstration of optical colorless direct-detection OFDM signals with 16- and 64-QAM formats beyond 15 Gb/s,” The 34th European Conference on Optical Communication (ECOC), (Sep. 2008), Paper Mo.3.E.1.
15. B. J. C. Schmidt, A. J. Lowery, and J. Armstrong, “Experimental demonstrations of electronic dispersion compensation for long-haul transmission using direct-detection optical OFDM,” J. Lightwave Technol. 26(1), 196–203 (2008). [CrossRef]
16. M. Schuster, S. Randel, C. A. Bunge, S. C. J. Lee, F. Breyer, B. Spinnler, and K. Petermann, “Spectrally Efficient Compatible Single-Sideband Modulation for OFDM Transmission With Direct Detection,” IEEE Photon. Technol. Lett. 20(9), 670–672 (2008). [CrossRef]
17. J. A. P. Morgado and A. V. T. Cartaxo, “Directly modulated laser parameters optimization for metropolitan area networks utilizing negative dispersion fibers,” IEEE J. Sel. Top. Quantum Electron. 9(5), 1315–1324 (2003). [CrossRef]