Open Access
Add to BibSonomyAbstract
Self-seeded passive optical networks (PONs) are currently attracting extensive research interest. In this paper, a novel self-seeded PON transmitter is, for the first time, proposed and experimentally demonstrated, which incorporates two face-to-face-positioned reflective semiconductor optical amplifiers (RSOAs) operating at their gain saturation regions: one RSOA directly driven by an upstream electrical signal and the other RSOA biased at a fixed current. Detailed experimental explorations are undertaken of the dynamic performance characteristics of the proposed transmitter. It is shown that, in comparison with previously reported self-seeded transmitters each employing a reflective mirror and a single electrical signal-driven RSOA, the proposed transmitter has a number of salient advantages including, considerably narrowed optical signal spectra, up to 16dB reduction in RINs of intensity-modulated optical signals, and residual intensity modulation crosstalk suppression as high as 10.7dB. The aforementioned features enable experimental demonstrations of real-time self-seeded 10Gb/s optical OFDM (OOFDM) transmitters. In particular, by making use of two low-cost RSOAs having their 3-dB modulation bandwidths as small as 1.125GHz, 10Gb/s over 25km adaptive OOFDM transmissions with power penalties of 0.6dB are experimentally achieved in the simple self-seeded IMDD PON systems.
© 2014 Optical Society of America
1. Introduction
To satisfy the unprecedented explosive end-users’ bandwidth requirement, optical orthogonal frequency division multiplexing (OOFDM) is regarded as one of the most promising candidate technologies for practical implementation in next-generation, high-speed passive optical networks (PONs) beyond time and wavelength division multiplexed (TWDM) PONs (NG-PON2) [1,2], since OOFDM has a large number of inherent and unique advantages including, for example, high spectral efficiency, excellent performance adaptability to component/system/network imperfections owing to its capabilities of adaptive bit and power loading, potential for providing cost-effective technical solutions due to the full utilization of rapid advances in modern digital signal processing (DSP) technology, and rich DSP-enabled transceiver performance self-awareness and networking functionalities [3]. In addition to all the aforementioned salient advantages, OOFDM can also considerably enhance the dynamic transceiver flexibility and reconfigurability for realizing elastic PONs, and maintain their compatibility with existing TDM PONs, as well as offer hybrid dynamic bandwidth allocation (DBA) in both the frequency and time domains [4].
In order to further improve the cost-effectiveness of an OOFDM WDM PON, for upstream transmission use can be made of a reflective intensity modulation-based color-agnostic transmitter at each individual optical network unit (ONU), where a coherent CW light externally seeded from the corresponding optical line terminal (OLT) is intensity-modulated by the reflective intensity modulator utilizing, for example, a reflective semiconductor optical amplifier (RSOA) [5], a reflective electro-absorption modulator (REAM) [6] or a reflective Fabry-Perot laser [7]. In such reflective ONU architecture, the provision of the externally seeded CW light source to each ONU requires the employment of an expensive multi-wavelength laser array in the OLT. The laser array can be substituted by an incoherent broadband light source. A slice of the broadband spectrum is modulated by each ONU for upstream transmission. However, such a spectrum-sliced solution is just capable of transmitting low signal capacities because the intensity-modulated upstream optical signals suffer from strong performance impairments caused by: a) amplified spontaneous emission (ASE) beat noises; b) relatively broad spectrum-enhanced chromatic dispersion (CD), and c) high relative intensity noise (RIN) [8].
To completely obviate the necessity of externally injecting a lightwave into each ONU, use can be made of a simple coherent-like narrowband light source created by self-seeding. It has been shown [9, 10] that a RSOA-based self-seeding scheme has potential of reducing all the above-mentioned challenges associated with the spectrum-sliced incoherent broadband light approach. In the self-seeding scheme, a resonant fibre cavity formed with a RSOA, an optical filter and a reflective mirror over a part of the PON transmission system, produces a self-tunable coherent-like narrowband light source, which is intensity-modulated and amplified simultaneously by the RSOA in the ONU to convey upstream data. For simplicity, throughout the paper, such a self-seeding scheme is referred to as the mirror scheme. So far, the mirror scheme has been experimentally demonstrated to be capable of supporting multi-gigabits non-return-to-zero on-off keying (NRZ-OOK) signal transmissions over relatively long distances [11–13]. More recently, by making use of a 4GHz modulation bandwidth RSOA in the mirror scheme, NRZ-OOK transmissions up to 10.7Gb/s have also been reported utilizing complicated electronic equalizations in the OLT [14].
Compared to conventional semiconductor laser diodes, the long self-seeded fibre cavity of the mirror scheme accommodates a large number of RSOA ASE noise-initiated longitudinal modes having random phases, this imposes a high RIN on the modulated optical signal. Furthermore, a portion of the already modulated optical signal in the cavity is also reflected, by the mirror, back into the RSOA, the reflected optical signal is then re-modulated by the RSOA employing the driving signal carrying different data patterns. Inevitably, this results in the residual intensity modulation crosstalk effect. Both the enhanced RIN and the residual intensity modulation crosstalk underpin the utilization of only conventional NRZ-OOK modulation formats of high signal extinction ratios in all those experimental demonstrations reported so far [9–14]. Given the fact that commercially available, low-cost RSOAs have very limited signal modulation bandwidths, to further improve the transmission performance of the self-seeded PON systems via fully exploiting the OOFDM advantages mentioned above, the adoption of highly spectral efficient OOFDM is significantly advantageous, which is, however, accompanied with low signal extinction ratio in intensity modulation and direct detection (IMDD) systems. Therefore, it is greatly beneficial if a novel RSOA-based self-seeding scheme capable of effectively reducing the aforementioned unwanted effects associated with the mirror scheme is proposed to enable the employment of OOFDM in the simple IMDD PON systems.
In this paper, a face-to-face dual-RSOA-based self-seeding scheme is, for the first time, proposed and experimentally demonstrated, in which the reflective mirror incorporated in the mirror scheme is replaced by a DC-biased RSOA, and the fibre cavity still consisting of an optical filter is formed with a signal-modulated RSOA and a DC-biased RSOA. Similarly, throughout the paper, the proposed scheme is termed the dual-RSOA scheme. Detailed experimental investigations show that, in comparison with the mirror scheme, the dual-RSOA scheme considerably narrows the bandwidth of the resulting coherent-like light source, reduces the RIN of the modulated optical signal by up to 16dB, and also suppresses the residual intensity modulation crosstalk by as high as 10.7dB. As a direct result, real-time self-seeded OOFDM transmitters are experimentally demonstrated, which allow adaptive modulation formats as high as 64-QAM to be adopted. In particular, based on two RSOAs with their small-signal 3-dB modulation bandwidths as narrow as 1GHz, 10Gb/s over 25km single-mode fiber (SMF) transmissions of adaptively loaded OOFDM signals are experimentally achieved with power penalties of 0.6dB in the self-seeded IMDD PON systems.
2. Proposed dual-RSOA self-seeding scheme and corresponding OOFDM IMDD PON system
The diagram of the proposed dual-RSOA self-seeding scheme is illustrated in the green-shaded area of Fig. 1. The scheme consists of two face-to-face RSOAs positioned at both ends of the fibre cavity: one signal-modulated RSOA (RSOA1) and the other DC-biased RSOA (RSOA2). Between RSOA1 and RSOA2, a tunable optical filter and a 2 × 1 optical coupler are also inserted in the fibre cavity. RSOA1 is directly driven by an electrical upstream signal to perform the electrical-to-optical conversion, whilst RSOA2 operates at its optical gain saturation region in order to achieve the following three desirable functionalities: a) narrowing the bandwidth of the self-seeding-created light source via its optical gain saturation dynamic mechanisms; b) reducing the modulated optical signal RIN arising from both the ASE-ASE/ASE-signal beat noises [15] and CD-induced conversion from frequency fluctuation to intensity fluctuation, and c) suppressing the residual intensity modulation crosstalk [16].
In practical implementation, RSOA1 is always positioned in the ONU, whilst the locations of RSOA2 and the corresponding optical filter and optical coupler are relatively flexible, which may be allocated in the remote node (RN), the OLT or the ONU. For the case of locating RSOA2 in the RN, in addition to their well-defined functionalities in TWDM NG-PON2, a standard array waveguide grating (AWG) can replace the required optical filter to perform the automatic wavelength selection, and a power splitter can also substitute the optical coupler to output one portion of the modulated optical signal for upstream transmission. Clearly, different RSOA2 locations give rise to different fibre cavity lengths of the dual-RSOA scheme. To highlight the dynamic performance characteristics of the proposed dual-RSOA scheme, throughout the paper, the case of positioning RSOA2 in the ONU is considered, which corresponds to a fixed fibre cavity length of 10m. It should be noted that our recent experimental results have shown that the dual-RSOA fibre cavity with a length of >1km can support 10Gb/s over 25km OOFDM IMDD transmissions. In our research lab, detailed experimental investigations are currently undertaken of the impact of various fibre cavity lengths on the upstream transmission performance of the dual-RSOA-based self-seeded PON systems, and results will be reported elsewhere in due course.
The considered entire OOFDM IMDD PON transmission system incorporating the proposed dual-RSOA scheme can be found in Fig. 1, and the adopted key transceiver/system parameter values are listed in Table 1. At the transmitter side, the real-time OOFDM transmitter is implemented with a field programmable gate array (FPGA) for high-speed DSPs and a 4GS/s digital-to-analogue converter (DAC) with 8-bits resolution. As detailed descriptions have already been reported of the transmitter’s DSP architecture in [17,18], an outline of the transmitter DSP design is, therefore, presented here. The core transmitter DSP functionalities include pseudo-random data generation, pilot-tone insertion, online adaptive bit and power loading for 15 data-carrying subcarriers with signal modulation formats selected from 16-quaternatry amplitude modulation (QAM), 32-QAM and 64-QAM, a 32-point inverse fast Fourier transform (IFFT) with input complex data being arranged to satisfy the Hermitian symmetry, online signal clipping level adjustment, 8-bit sample quantization and cyclic prefix (CP) addition. The above OOFDM transmitter design offers live optimisations of subcarrier bit/power allocation, and digital signal clipping level using the FPGA embedded memory editor via a joint test action group (JTAG) connection to a personal computer. Such parameter manipulation capability enables not only the rapid identification of optimum system parameters but also the examination of the highest achievable system performances for different system configuration scenarios.
Table 1. OOFDM Transceiver and System Parameters
As shown in Fig. 1, the digital OFDM signal emerging from the FPGA is first converted to an analog electrical signal via the DAC, and the converted signal power level is then optimized by variable electrical attenuators and a 4.2GHz RF amplifier. The resulting analog electrical OFDM signal is combined with an 80mA DC bias current in a 6GHz bias tee to directly modulate ROSA1 operating at a temperature of 13°C. Such a temperature is not critical for obtaining the system performances presented in the following sections of the paper. RSOA1 amplifies and modulates the optical signal of a spectrum initially selected by a tunable optical bandpass filter (OBPF). The RSOA1-modulated optical signal is fed to a 50/50 optical coupler. The optical coupler split the optical signal into two portions: one portion first passes through a polarization controller (PC), then it is amplified and reflected again by RSOA2 and finally fed to RSOA1 for remodulation; whilst the other portion is launched into a 25km SMF IMDD transmission system without inline optical amplification and dispersion compensation. Here the PC is employed to adjust the optical signal polarization state to align with the high gain polarization direction of polarization-sensitive RSOA2 (RSOA1 is polarization-insensitive). The utilization of the PC is, however, not necessary when polarization-insensitive RSOA2 is employed. It is also envisaged that a combined pair of a RSOA and a Faraday rotator allows the use of any two RSOAs in the proposed scheme regardless their polarization sensitivities [14].
At the receiver side, the optical signal passes through a variable optical attenuator to adjust the received optical power (ROP) and is subsequently directly detected by a 12GHz p-i-n detector to convert the OOFDM signal into the electrical domain. After having been filtered by a 2.2GHz low pass filter, the received analog electrical OFDM signal is captured and digitized by a real-time oscilloscope and then processed using Matlab. Similar to those reported in [17,18], the receiver DSP procedures include symbol synchronization, pilot-subcarrier detection, channel estimation/equalization, and all other functions that are just inverse to their transmitter counterparts. The bit error rates (BERs) of each individual subcarrier and the entire OFDM channel are continuously calculated and displayed in the oscilloscope. This enables the rapid optimization of the overall transmission systems via system parameter adjustments including adaptive subcarrier bit and power loading.
3. Experimental results
To explicitly exhibit the advantages of the proposed dual-RSOA scheme compared to the mirror scheme, in Subsections 3.1-3.3, extensive experimental explorations are first undertaken of the dynamic optical spectral characteristics of the self-seeding-generated light sources, RSOA2 gain saturation-dependent reductions in both the RINs and residual intensity modulation crosstalk of the modulated optical signals. An in-depth understanding of these physical mechanisms paves a solid path leading to the experimental demonstrations of 10Gb/s over 25km SMF adaptive OOFDM transmissions in the dual-RSOA scheme-based IMDD PON systems in Subsection 3.4.
3.1 Narrowed optical spectral width
For different RSOA2 bias currents, the optical spectra of various self-seeding-generated lights emerging from the 50/50 optical coupler at point “A” indicated in Fig. 1 are shown in Fig. 2, where the output optical spectrum of the mirror scheme and the measured frequency response of the adopted OBPF are also plotted for comparisons. In obtaining Fig. 2, RSOA1 is biased at a fixed current of 80mA (no OFDM driving currents applied to RSOA1) to highlight the gain saturated RSOA2-induced optical spectral dynamics only. In measuring the characteristics of the mirror scheme, RSOA2 is replaced by a reflective mirror without introducing any changes to all other components and their operating parameters.
Fig. 2 Measured output optical spectra of the dual-RSOA self-seeding scheme for different RSOA2 bias currents. The corresponding output optical spectrum of the mirror scheme is also shown together with the measured frequency response of the adopted OBPF. RSOA1 is biased at a fixed current of 80mA.
It can be seen in Fig. 2 that, for both the dual-RSOA and mirror schemes, their output optical spectral widths are considerably narrower than the OBPF bandwidth, and that their dominant spectral peaks are red-shifted with respect to the center wavelength of the OBPF frequency response. These behaviors agree well with experimental results reported recently in [19]. The physics underpinning the observed red-shifted spectral peaks is the Bogatov effect [20], which describes the process of how red-shifted dominant spectral components are generated due to optical amplification-associated cross gain modulation arising from the interactions between multiple longitudinal modes. The spectral red-shifting implies that, in comparison with short-wavelength modes, higher optical gains are experienced by long-wavelength modes, which, however, also suffer from larger attentions introduced by the long-wavelength edge of the OBPF frequency response. The co-existence of the above two dynamics determines that the red-shifted main spectral peaks always occur at the long-wavelength edge of the OBPF frequency response [21].
It should be noted, in particular, in Fig. 2 that, in comparison with the mirror scheme, the dual-RSOA scheme can further suppress the optical spectral widths and simultaneously reduce the corresponding sidelobes. Such suppression effect is enhanced for higher RSOA2 bias currents. This indicates that RSOA2 operating at its gain saturation region reinforces the longitudinal mode interactions within the self-seeded cavity, thus leading to fewer survived modes and subsequently much narrower signal spectral widths produced. Here it is also worth pointing out the following three aspects: a) a narrow spectral width of the self-seeding-generated light not only improves the tolerance of the scheme to CD impairments but also contributes to the RIN reduction, as discussed in Section 3.2. b) in comparison with the OBPF frequency response width, the considerably narrowed signal spectrum clearly indicates the coherent-like nature of the dual-RSOA self-seeding-generated lightwaves, and c) if the shape of the OBPF frequency response can be engineered appropriately, then the unwanted linear cavity effects such as CD can be compensated using the OBPF.
3.2 Reduced RIN
For both the dual-RSOA and mirror schemes, Fig. 3(a) shows the RIN spectra measured for various RSOA2 bias currents. For simplicity, for a specific frequency, the RIN reduction is defined as the ratio between the RIN level of the dual-RSOA scheme and that corresponding to the mirror scheme, the RSOA2 bias current-dependent RIN reduction is shown in Fig. 3(b) for three representative frequencies of 0.5GHz, 1GHz and 2GHz, which are located at the low, middle and high frequencies of the OFDM signal spectrum. In obtaining Figs. 3(a) and 3(b), the RSOA1 operating conditions are identical to those adopted in Fig. 2.
Fig. 3 (a) Measured RIN spectra for both the dual-RSOA and mirror self-seeding schemes; (b) RSOA2 bias current-dependent RIN reduction for the dual-RSOA self-seeding scheme at representative frequencies. In measuring both (a) and (b), the RSOA1 bias current is fixed at 80mA.
It is shown in Figs. 3(a) and 3(b) that, with respect to the mirror scheme, the dual-RSOA scheme can significantly lower the RIN level especially in the low frequency region, and the RIN reduction frequency region broadens rapidly with increasing RSOA2 bias current. For RSOA2 bias currents above 65mA, the RIN reductions as high as 16dB are obtainable across the whole signal spectral region. The observed RIN reductions are because of the following two effects: a) optical spectral narrowing presented in Subsection 3.1, and b) RSOA2-induced increase in optical power in the fibre cavity. On the other hand, the broadening of the RIN reduction frequency region is a direct result of high-pass filtering associated with gain-saturated RSOA2.
3.3 Suppressed residual intensity modulation crosstalk
To demonstrate the effectiveness of the dual-RSOA scheme in reducing the residual intensity modulation crosstalk effect, the OOFDM waveforms of identical optical powers of −7dBm are plotted in Fig. 4(a), where the experimental measurements are undertaken at point “A” and point “B”, as indicated in the experimental system setup of Fig. 1. Here point “B” corresponds to an output port of a 90/10 optical coupler that is newly introduced only for the measurements in this subsection. The optical waveform measured at point “A” represents the optical signal input to RSOA2 and the optical waveform measured at point “B” represents the optical signal emerging from RSOA2. Both the driving voltage and bias current of RSOA1are given in Table 1 and the bias current of RSOA2 is fixed at 95mA.
Fig. 4 (a) Comparisons of OOFDM signal waveforms before and after passing through RSOA2 biased at a DC current of 95mA; (b) Residual intensity modulation crosstalk suppression versus bias current of RSOA2.
It can be seen in Fig. 4(a) that, compared to the optical signal input to RSOA2, the intensity fluctuations of the output optical signal is significantly suppressed. Such fluctuation suppression is a direct result of RSOA2 self-gain modulation: high signal peaks experience low optical gains and deep signal nulls see high optical gains, thus producing smoothed signal waveforms after RSOA2 [22].
By considering the noise-like nature of typical OOFDM waveforms [23] and the conventional concept of optical modulation amplitude (OMA) for NRZ/RZ signal modulation formats, in order to explicitly quantify the above-observed residual intensity modulation crosstalk reduction effect, the OOFDM OMA is defined as:
where is the average optical power, () is the optical power of the ith (jth) signal sample with its value satisfying (), is the sampling duration, () is the number of samples satisfying () within the entire OOFDM sample sequence captured, and is the total number of samples. Based on Eq. (1), the residual intensity modulation crosstalk suppression (RCS) can be written aswhere () is the OMA of the OOFDM signal input to (output from) RSOA2. Based on Eq. (3), the corresponding OMAs of the input and output optical signals shown in Fig. 4(a) are 0.0317mW and 0.0027mW, respectively, this gives rise to a RCS of 10.7dB. As an increase in RSOA2 bias current speeds up the RSOA2 gain dynamic responses, this leads to an increased cut-off frequency of the high-pass frequency response of RSOA2, and subsequently larger RSOA2 bias current-enhanced RCSs. The analyses are verified in Fig. 4(b), which shows that the RCS (in dB) is almost proportional to RSOA2 bias current.3.4 Improved system transmission performance
In this subsection, full use is made to all the above-discussed dynamic features of the dual-RSOA scheme to experimentally demonstrate 10Gb/s over 25km SMF OOFDM transmissions in the IMDD PON systems. In addition, to identify key factors limiting the maximum achievable system performance, comparisons are also made between the dual-RSOA and mirror schemes for various system architectures.
To effectively compensate for the overall system frequency response roll-off effect [5] and simultaneously maximize the system transmission capacity, parameter optimizations are first conducted via adaptive bit and power loading on the information-bearing subcarriers. For the dual-RSOA scheme-based 10Gb/s@25km SMF IMDD PON system with RSOA2 biased at 95mA, the adaptively loaded and received subcarrier power profiles and measured system frequency responses are given in Fig. 5(a), where each system frequency response is measured from the transmitter IFFT input to the receiver FFT output and normalized to the first subcarrier power. It can be found in Fig. 5(a) that, for high frequency subcarriers, the maximum system frequency response roll-off is approximately 21dB, which agrees with that reported in [5]. On the other hand, the resulting optimum subcarrier bit allocation profile is shown in Fig. 5(b), based on which the raw signal bit rate of 10Gb/s can be easily worked out utilizing the transceiver parameters listed in Table 1. It is shown in Fig. 5(b) that low signal modulation formats occur on high frequency subcarriers, this is mainly attributed to the limited FPGA dynamic range-induced residual system frequency response roll-off effect, as seen in Fig. 5(a).
Fig. 5 (a) Adaptively loaded and received subcarrier power profiles and normalized system frequency response; (b) Adaptive subcarrier bit allocation profiles. In these two Figs, different transmission system configurations are considered including the 10Gb/s dual-RSOA scheme and the 3Gb/s mirror/dual-RSOA schemes.
By making use of the optimum power and bit loading profiles presented in Figs. 5(a) and 5(b), Fig. 6(a) shows the measured BER performances of 10Gb/s adaptive OOFDM signal transmissions over the dual-RSOA scheme-based IMDD PON systems for both the optical back-to-back (BTB) and 25km SMF system configurations. As seen in Fig. 6(a), after 25km SMF transmission, the minimum total channel BER as low as 4.3 × 10−4 is obtainable at a ROP of −3dBm, equally importantly, the observed power penalty at the forward error correction (FEC) limit of 2.3 × 10−3 [24] is approximately 0.6dB, which is very similar to that published in [5], where 1GHz RSOA intensity-modulated 7.5Gb/s over 25km SMF OOFDM IMDD transmissions are experimentally demonstrated using an externally seeded coherent CW light source. The power penalty similarity indicates, once again, the coherent-like nature of the dual-RSOA scheme-generated optical signal, since an optical signal of a narrowed spectral bandwidth enhances the system tolerance to various impairments associated with fibre CD. It can also be seen in Fig. 6(a) that the 10Gb/s over 25km OOFDM IMDD transmission system discussed above has a loss budget of approximately 18dB assuming a 9dBm optical signal power at the transmitter. The loss budget is insufficient to support standardized PON systems. However, the loss budget can be improved by at least 10dB when use is made of: a) an APD instead of the PIN [25], b) RSOA1 and RSOA2 with higher optical output saturation powers and c) reduced fibre cavity losses by further cavity optimizations.
Fig. 6 (a) BER performances of 10Gb/s OOFDM IMDD transmissions in the dual-RSOA scheme-based self-seeded PON architectures: optical BTB and 25km SMF. (b) Corresponding received constellations of representative subcarriers before and after performing channel equalization after the 25km SMF transmission at a ROP of −3dBm. For both (a) and (b), RSOA1 and RSOA2 are biased at 80mA and 95mA, respectively.
Before and after performing channel equalization in the receiver, the corresponding constellations of representative subcarriers are exemplified in Fig. 6(b), which are recorded at a ROP of −3dBm in the dual-RSOA scheme-based 10Gb/s over 25km SMF IMDD system. As expected from Fig. 5(a), for all these constellations measured prior to channel equalization, large variations in subcarrier amplitude occur in Fig. 6(b), which can, however, be effectively rectified using channel equalization.
To further examine the effectiveness of the dual-RSOA scheme in minimizing the unwanted effects, via simply substituting RSOA2 with a reflective mirror and maintaining all other system components and their parameters unchanged, the dual-RSOA scheme is replaced by the mirror scheme. After extensive system optimizations via adaptive bit and power loading, it is found that the mirror scheme is just capable of supporting a maximum signal bit rate as low as 3Gb/s in the 25km SMF IMDD PON system. To detail the transmission performance of the mirror scheme-based case, the resulting optimum subcarrier power and bit loading profiles are presented in Figs. 5(a) and 5(b), respectively. As seen in Fig. 5(b), only 5 low frequency subcarriers are employable to achieve a total channel BER below the FEC limit. The BER performances of the mirror scheme-based 3Gb/s OOFDM transmissions are plotted in Fig. 7(a) for both the BTB and 25km SMF system configurations. On the other hand, Fig. 7(b) illustrates the equalized constellations of the 1st, 3rd, and 5th subcarriers at a ROP of −12dBm.
Fig. 7 (a) BER performances of 3Gb/s OOFDM signals based on the dual-RSOA and mirror schemes for the optical BTB and 25km SMF system configurations. RSOA1 and RSOA2 are biased at 80mA and 95mA, respectively; (b) Received constellations of different subcarriers after performing channel equalization for the 25km SMF transmission (ROP = −12dBm).
It is clear from Fig. 7(a) that, for the mirror scheme, error floors corresponding to BERs of approximately 2.0 × 10−3 begin to occur for ROPs as low as −10dBm. Such BER developing trends indicate that the strong effects of both RIN and residual intensity modulation crosstalk play significant roles in determining the maximum achievable performance of the mirror scheme. Moreover, the corresponding power penalty at the FEC limit is about 3dB, which is significantly higher than that observed in Fig. 6(a). This is because the wide signal spectrum associated with the mirror scheme considerably decreases the system tolerance to all impairments induced by CD.
Furthermore, by making use of the identical 3Gb/s bit and power loading profiles optimised for the mirror scheme-based case, a simple substitution of the mirror with RSOA2 results in significant BER reductions to values well below 1.0 × 10−4 at a ROP of −13dBm, and also brings about a negligible power penalty, as shown in Fig. 7(a). This suggests that the observed performance improvements are attributed by the dual-RSOA scheme rather than the adaptive bit and power loading capability.
4. Conclusions
A novel self-seeding scheme incorporating two face-to-face optical gain-saturated RSOAs has been proposed and experimentally demonstrated, for the first time, for use in ONUs of OOFDM IMDD PON systems. Detailed experimental investigations have been undertaken of the dynamic performance characteristics of the proposed scheme. It has been shown that, in comparison with the previously published mirror scheme, the proposed scheme has a number of salient advantages including, considerably narrowed optical signal spectra, up to 16dB reduction in RINs of intensity-modulated optical signals, and residual intensity modulation crosstalk suppression as high as 10.7dB. The aforementioned features enable experimental demonstrations of real-time self-seeded OOFDM transmitters for upstream transmissions. In particular, by making use of two low-cost 1GHz RSOAs, 10Gb/s over 25km adaptive OOFDM transmissions with power penalties of 0.6dB have been experimentally achieved in simple self-seeded IMDD PON systems.
Acknowledgments
This work was supported in part by the PIANO + under the European Commission’s ERA-NET Plus Scheme within the project OCEAN under Grant Agreement 620029, in part by Sino-UK Higher Education Research Partnership for PhD Studies, National High Technology Research and Development Program of China (863 Program) (2012AA011302, 2012AA011304, 2013AA010503), NSFC (No. 61071097, No. 61107060, No. 61101095), the Fundamental Research Funds for the Central Universities of China (ZYGX2011J009).
References and links
1. E. Wong, “Next-generation broadband access networks and technologies,” J. Lightwave Technol. 30(4), 597–608 (2012). [CrossRef]
2. N. Cvijetic, “OFDM for next generation optical access networks,” J. Lightwave Technol. 30(4), 384–398 (2012). [CrossRef]
3. R. Giddings, “Real-time digital signal processing for optical OFDM-based future optical access Networks,” J. Lightwave Technol. 32(4), 553–570 (2014). [CrossRef]
4. N. Cvijetic, D. Qian, and J. Hu, “100 Gb/s optical access based on optical orthogonal frequency-division multiplexing,” IEEE Commun. Mag. 48(7), 70–77 (2010). [CrossRef]
5. R. P. Giddings, E. Hugues-Salas, X. Q. Jin, J. L. Wei, and J. M. Tang, “Experimental demonstration of real-time optical OFDM transmission at 7.5 Gb/s over 25-km SSMF using a 1-GHz RSOA,” IEEE Photon. Technol. Lett. 22(11), 745–747 (2010). [CrossRef]
6. E. Hugues-Salas, R. P. Giddings, X. Q. Jin, Y. Hong, T. Quinlan, S. Walker, and J. M. Tang, “REAM intensity modulator-enabled 10Gb/s colorless upstream transmission of real-time optical OFDM signals in a single-fiber-based bidirectional PON architecture,” Opt. Express 20(19), 21089–21100 (2012). [CrossRef] [PubMed]
7. Z. Xu, Y. Yeo, X. Cheng, and E. Kurniawan, “20-Gb/s injection locked FP-LD in a wavelength-division-multiplexing OFDM-PON,” Optical Fiber Communication/National Fiber Optic Engineers Conference (OFC/NFOEC), (Los Angeles, 2012), Paper OW4B.3. [CrossRef]
8. K. H. Han, E. S. Son, H. Y. Choi, K. W. Lim, and Y. C. Chung, “Bidirectional WDM PON using light-emitting diodes spectrum-sliced with cyclic arrayed-waveguide grating,” IEEE Photon. Technol. Lett. 16(10), 2380–2382 (2004). [CrossRef]
9. E. Wong, K. L. Lee, and T. B. Anderson, “Directly modulated self-seeding reflective semiconductor optical amplifiers as colorless transmitters in wavelength division multiplexed passive optical networks,” J. Lightwave Technol. 25(1), 67–74 (2007). [CrossRef]
10. L. Marazzi, P. Parolari, R. Brenot, G. de Valicourt, and M. Martinelli, “Network-embedded self-tuning cavity for WDM-PON transmitter,” Opt. Express 20(4), 3781–3786 (2012). [CrossRef] [PubMed]
11. T. Komljenovic, D. Babić, and Z. Sipus, “47-km 1.25-Gbps transmission using a self-seeded transmitter with a modulation averaging reflector,” Opt. Express 20(16), 17386–17392 (2012). [CrossRef] [PubMed]
12. Q. Deniel, F. Saliou, P. Chanclou, D. Erasme, and R. Brenot, “Up to 45km-long amplified self-seeded RSOA based external cavity for 2.5Gb/s WDM PON transmission,” Optical Fiber Communication /National Fiber Optic Engineers Conference (OFC/NFOEC), (Anaheim, 2013), Paper OW4D.2. [CrossRef]
13. Q. Deniel, F. Saliou, S. D. Le, P. Chanclou, D. Erasme, and R. Brenot, “Amplified RSOA self-tuning laser for WDM PON using saturated SOA for noise reduction and data cancellation,” European Conference on Optical Communication (ECOC), (London, 2013), Paper We1.F.5. [CrossRef]
14. L. Marazzi, P. Parolari, M. Brunero, A. Gatto, M. Martinelli, R. Brenot, S. Barbet, P. Galli, and G. Gavioli, “Up to 10.7-Gb/s High-PDG RSOA-based colorless transmitter for WDM networks,” IEEE Photon. Technol. Lett. 25(7), 637–640 (2013). [CrossRef]
15. X.-H. Jia, Z.-M. Wu, and G.-Q. Xia, “Detailed theoretical investigation on relative intensity noise reduction enhancement based on reflective SOAs,” Opt. Laser Technol. 44(5), 1240–1246 (2012). [CrossRef]
16. S. O. Dúill, L. Marazzi, P. Parolari, R. Brenot, C. Koos, W. Freude, and J. Leuthold, “Efficient modulation cancellation using reflective SOAs,” Opt. Express 20(26), B587–B594 (2012). [CrossRef] [PubMed]
17. R. P. Giddings, X. Q. Jin, E. Hugues-Salas, E. Giacoumidis, J. L. Wei, and J. M. Tang, “Experimental demonstration of a record high 11.25Gb/s real-time optical OFDM transceiver supporting 25km SMF end-to-end transmission in simple IMDD systems,” Opt. Express 18(6), 5541–5555 (2010). [CrossRef] [PubMed]
18. X. Jin, J. Wei, R. Giddings, T. Quinlan, S. Walker, and J. Tang, “Experimental demonstrations and extensive comparisons of end-to-end real-time optical OFDM transceivers with adaptive bit and/or power loading,” IEEE Photonics Journal 3(3), 500–511 (2011). [CrossRef]
19. F. Xiong, W.-D. Zhong, M. Zhu, H. Kim, Z. Xu, and D. Liu, “Characterization of directly modulated self-seeded reflective semiconductor optical amplifiers utilized as colorless transmitters in WDM-PONs,” J. Lightwave Technol. 31(11), 1727–1733 (2013). [CrossRef]
20. A. P. Bogatov, P. G. Eliseev, and B. N. Sverdlov, “Anomalous interaction of spectral modes in a semiconductor laser,” IEEE J. Sel. Top. Quantum Electron. 11(7), 510–515 (1975). [CrossRef]
21. S. L. Girard, M. Piche, H. Chen, G. W. Schinn, W.-Y. Oh, and B. E. Bouma, “SOA fiber ring lasers: single-versus multiple-mode oscillation,” IEEE J. Sel. Top. Quantum Electron. 17(6), 1513–1520 (2011). [CrossRef]
22. K. Sato and H. Toba, “Reduction of mode partition noise by using semiconductor optical amplifiers,” IEEE J. Sel. Top. Quantum Electron. 7(2), 328–333 (2001). [CrossRef]
23. J. M. Tang and K. A. Shore, “30-Gb/s signal transmission over 40-km directly modulated DFB-laser-based single-mode-fiber links without optical amplification and dispersion compensation,” J. Lightwave Technol. 24(6), 2318–2327 (2006). [CrossRef]
24. X. Q. Jin, E. Hugues-Salas, R. P. Giddings, J. L. Wei, J. Groenewald, and J. M. Tang, “First real-time experimental demonstrations of 11.25Gb/s optical OFDMA PONs with adaptive dynamic bandwidth allocation,” Opt. Express 19(21), 20557–20570 (2011). [CrossRef] [PubMed]
25. Q. W. Zhang, E. Hugues-Salas, Y. Ling, H. B. Zhang, R. P. Giddings, J. J. Zhang, M. Wang, and J. M. Tang, “Record-high and robust 17.125 Gb/s gross-rate over 25 km SSMF transmissions of real-time dual-band optical OFDM signals directly modulated by 1 GHz RSOAs,” Opt. Express 22(6), 6339–6348 (2014). [CrossRef] [PubMed]
