Open Access
Add to BibSonomy
Abstract
In this paper, we experimentally demonstrate the transmission of a 100 Gb/s/λ PAM-4 signal over a 40/80-km single mode fiber (SMF) in the O-band utilizing a 4-bit resolution digital-to-analog converter (DAC) for signal generation. Low resolution DACs are preferred to meet the requirement of low-cost criteria of datacenter interconnects (DCIs). However, large quantization noise introduced by low resolution DACs will deteriorate system performance significantly. Noise shaping (NS) technique is investigated to reduce the quantization noise within the PAM-4 signal band. The experimental results show that the bit error ratio (BER) performance of the signal generated by 4-bit resolution DAC and NS technique will approach that of the signal generated by the 8-bit resolution DAC in the 40/80-km optical fiber transmission system of a 50 Gbaud PAM-4 signal in the O-band, which indicates that our proposed scheme operating in the O-band with a 4-bit resolution DAC and NS technique is a promising candidate for 100 Gbit/s/λ beyond a 40-km Inter-DCI.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
The popularity of network and computers has brought us into the era of information and technological innovation. Many kinds of high broadband services, such as social media (SM), cloud computing (CC) and high-division television (HDTV), are emerged to satisfy the demand of people for information interaction, which results in the rapid growth of transmission capacity requirement for datacenter interconnects (DCIs) [1–6]. Solutions for 100 Gb/s and 400 Gb/s signals over 10-km single mode fiber (SMF) transmission are standardized in IEEE 802.3cu-2021, in which 100 Gb/s/λ transmission rate is implemented [5]. However, the distance of Inter-DCI has exceeded to 40-km even 80-km, and 100 Gb/s/λ signal for beyond 40-km Inter-DCI is still an important topic and needs to be further studied.
Numerous works have been reported to achieve 100 Gb/s/λ beyond 40-km SMF transmission, and most of them are realized by IQ modulator and coherent receiver in C-band [6–11], since the conventional intensity modulation and direct detection (IM/DD) system will suffer serious frequency selective fading (FSF) caused by chromatic dispersion (CD) [12–15]. Due to the negligible dispersion characteristic of O-band system, transmission of 100 Gb/s/λ signal in O-band IM/DD system is an alternative scheme, which can be used to achieve 20-km SMF transmission or even beyond [16–23]. Researches on 100 Gb/s/λ signal transmission in O-band mainly focus on the elimination of signal distortions induced by bandwidth limitation and nonlinearity cancellation through various digital signal processing (DSP) techniques [16–20]. D. Zou et al. experimentally demonstrated the transmission of 95 Gb/s Discrete Multi-tone (DMT) signal over 20-km SMF in O-band IM/DD system, implementing optimized Quadric Curve (QC) fitting-based pre-equalization to eliminate the influence of bandwidth limitation and discrete-Fourier-transform spread (DFT-Spread) for peak-to-average power ratio (PAPR) reduction [16]. F. Li et al. reported the transmission of 100 Gb/s/λ PAM-N signal over 40-km SMF in IM/DD system with 10-G class direct modulation laser (DML) in O-band, and look-up-table (LUT) is used to compensate the nonlinear distortions [17,18]. While the transmission distance requirement is extended to 50-km and 80-km, semiconductor optical amplifier (SOA) is necessary for transmission link power loss compensation [21–23]. For all afore-mentioned systems, 8-bit resolution digital-to-analog converters (DACs) are used for digital-to-analog conversion in the transmitter. In order to meet the requirement of low-cost criteria of Inter-DCI, signal generated with low resolution DACs is a feasible solution to effectively reduce the system cost. However, quantization noise problem arises when low-resolution DAC is adopted for signal generation in the DCI, thus techniques for quantization noise suppression need to be explored and studied.
Noise shaping (NS) technique, delta-sigma modulation (DSM) and digital resolution enhancer (DRE) have been proposed to eliminate the in-band quantization noise. However, DSM requires a quite high oversampling ratio (OSR=8∼16) in the signal transmission, and the computational complexity of DRE is high due to the implementation of Viterbi algorithm [24–28]. Thus, DSM and DRE techniques are not suitable for low-cost DCI. NS technique with very low computational complexity was firstly proposed in Ref. [24], with far less required OSR compared to DSM. However, NS technique has only been explored in DMT systems by simulation, the techniques needs to be further investigated in the practical transmission system where other types of noises except quantization noise also exist. Although bit and power loading technique provides an effective solution to cope with the FSF induced by CD in C-band for DMT, it is still very challenge to achieve 100Gbit/s/λ DMT signal over beyond 40-km SMF transmission [29]. In addition, DMT signal is vulnerable to system nonlinear distortions. Compared to DMT, pulse amplitude modulation (PAM) format is widely used in O-band over beyond 40-km SMF transmission for Inter-DCI [17,18,21–23]. The research of NS technique in transmission of PAM signal generated by low resolution DAC in O-band needs to be theoretically studied and experimentally verified to meet the requirement of low-cost DCIs.
In this paper, which is an extension of our previous work report in Ref. [25], the principle of NS technique is analyzed from the perspective of time-domain and frequency-domain, and the approach to obtain the feedback filter taps is also shown in this paper. In order to show the low-cost criterion of NS technique, the needed number of taps of feedback filter is simulated for 50/56/60 Gbaud PAM-4 signal transmission. Finally, we also experimentally demonstrate the transmission of 100 Gb/s/λ PAM-4 signal for realizing low-cost 80-km Inter-DCI using 4-bit resolution DAC in O-band assisted by NS technique for the first time. A SOA and an optical band-bass filter (OBPF) are used for 80-km SMF transmission at the receiver side to compensate the power loss and minimize the out-of-band amplified spontaneous emission (ASE) noise, respectively. To cope with the DAC clock leakage appearing at ±20 GHz in our experiment, a 2-taps adaptive notch filter (ANF) proposed in our previous work in [30] is adopted to eliminate this narrowband interference. A 1.5-dB receiver sensitivity improvement can be obtained for 56 Gbaud PAM-4 signal at the hard decision forward error correction (HD-FEC) threshold in optical back-to-back (OBTB) when ANF is applied to suppress the clock leakage. Based on above-mentioned techniques, BERs of 50 Gbaud PAM-4 signal generated by 4-bit resolution DAC can successfully reach HD-FEC after implementing NS technique over 40/80-km transmission in O-band. The BER performance of 50 Gbaud PAM-4 generated by 4-bit resolution DAC and NS technique will approach that of 50 Gbaud PAM-4 generated by the 8-bit resolution DAC.
The rest of this paper is organized as follows. Section 2 introduces the principle of NS. Experiment setups and results are described in Section 3. Finally, the summarization of this paper is in Section 4.
2. Noise shaping technique
The principle of NS technique is introduced in section 2.1, and the optimization number of taps of FIR described in section 2.1 is simulation tested in section 2.2.
2.1 Principle of the noise shaping technique
The rapid growth in the number of DCIs push the vendors to further reduce the system implementation costs [31]. High resolution DACs are not suitable for low-cost DCI. However, large quantization noise is inevitably introduced with low resolution DAC, which will deteriorate system performance significantly. The quantization noise can be modeled as uniformly distributed white noise, and introduced in the transmitter during digital-to-analog conversion process [24,25]. Figure 1(a) shows the structure of NS technique, and Q represents a DAC. The quantization noise $N({e^{j\omega }})$ is the difference between the data before and after DAC. And the output $Y({e^{j\omega }})$ can be expressed as:
where $H({e^{j\omega }})$ and $X({e^{j\omega }})$ represent the channel response of feedback filter and the signal before quantization, respectively.
Fig. 1. (a) The architecture of NS technique and (b) operating principle of NS technique in time-domain and (c) operating principle of NS technique in frequency-domain and (d) simulated quantization noise spectra of 56Gbaud PAM-4 signal generated by 4-bit DAC with and without NS technique.
In order to minimize the quantization noise within signal band, the quantization noise using NS technique should less than the quantization signal before this operation, just shown as below [24].
Suppose the feedback filter is a finite impulse response (FIR) filter with the form of:
Equation (5) can be rewritten as:
The operating principle of NS technique is to make transmitted signal close to DAC output levels by adding noise to the unused band of signal, and the out-band noise is obtained from the output of a FIR, where the input is the difference between the data before and after DAC last sample time point as shown in Fig. 1(a). Figure 1(b) gives the time-domain 56 Gbaud PAM-4 signal generated with 3-bit DAC with and without adding out-band noise, respectively. We can find that the quantization noise can be effectively reduced when the out of band noise is appropriately added. The operating principle of NS technique can also be described in frequency-domain. With the help of NS technique, quantization noise is mostly distributed outside of the signal band with only a small amount of quantization noise remained in signal band as shown in Fig. 1(c) and 1(d). Figure 1(c) gives the schematic diagram of electrical spectra of input signal $X({e^{j\omega }})$, the quantized output $Y({e^{j\omega }})$ without NS technique and the quantized output $Y({e^{j\omega }})$ with NS technique, respectively. And Fig. 1(d) gives the simulated quantization noise spectra of 56Gbaud PAM-4 signal generated by 4-bit DAC with and without NS technique.
2.2 Optimization taps of FIR
The principle of NS technique is to make transmitted signal close to output levels of DAC by adding noise to the unused band of signal. The noise is obtained from the output of a FIR as shown in Fig. 1(a). It is necessary to find the optimal number of taps of the FIR filter to balance in-band quantization noise elimination and computation complexity. In order to obtain the optimal number of taps of FIR filter, the SNR enhancement versus number of taps is calculated by simulation under different signal bandwidths, and the results are shown in Fig. 2.
As shown in Fig. 2, the SNR enhancement of quantization noise increases with number of taps increased, and no obvious improvement is observed when number of taps reaches to 7. As the signal bandwidth increases, the performance of the NS technique deteriorates due to the decreasing of the unused bandwidth. And the optimized number of taps used for further 50 Gbaud, 56 Gbaud, 60 Gbaud PAM-4 signal transmission experiment are 9, 8, 8, respectively.
3. Experimental setup and results
In this section, we describe the experimental setup in section 3.1. Section 3.2 and section 3.3 show the experiment results of 50 Gbaud, 56 Gbaud and 60 Gbaud PAM-4 signals based on ANF and NS technique.
3.1 Experimental setup
The proof-of-concept experimental setup and DSP diagrams of 100 Gb/s/λ PAM-4 signal over 40/80-km SMF transmission in IM/DD system are given in Fig. 3. Pseudo random binary sequence (PRBS) is firstly mapped into PAM-4 symbols, and training sequences are added in the front of payload PAM-4 symbols for receiver side synchronization and channel estimation. Digital pre-equalization using 21-taps FIR is adopted to eliminate the influence of inter-symbol-interference (ISI) induced by bandwidth limitation. Then the pre-equalized signal is reshaped by a root raised cosine (RRC) filter with roll-off factor of 0.125. The shaped signal is re-sampled to 80-GSa/s sampling rate, then NS technique is adopted to make signal close to DAC output levels by adding out-band noise. After adding out-band noise, signal is quantized to 16 different levels in MATLAB to simulate a 4-bit resolution DAC. And then, the generated signal is uploaded to an 80-GSa/s sampling rate Fujitsu DAC with 16.7 GHz 3-dB bandwidth and 8-bit resolution. The out-of-band quantization noise can be effectively suppressed due to the bandwidth limitation of DAC. If the end-to-end channel bandwidth is high, a low-pass filter (LPF) is needed to place after DAC to improve the performance of NS technique. An electrical amplifier (EA) with 50 GHz bandwidth and 23-dB gain is applied to boost the transmitted signal. A distributed feedback laser (DFB) with 10-dBm output optical power operating at 1310.43-nm is used to generate optical carrier signal. A single-arm Mach-Zehnder modulator (MZM) with 40 GHz 3-dB bandwidth is biased at the quadrature point to realize PAM-4 signal modulation and the output signal power is 5.8-dBm. In our experiment, to satisfy the requirement of Inter-DCI, the lengths of SMF are set as 40/80-km. The optical signal power after 40/80-km SMF transmission are approximately −8-dBm and −20.4-dBm, respectively. Since the BER of received signal with −20.4-dBm received optical power (ROP) cannot reach below the 7% HD-FEC threshold, a SOA with the gain of 25-dB and the noise figure <7-dB and an OBPF are used at the receiver side for signal amplification and out-of-band ASE noise elimination for the 80-km SMF transmission, respectively. At the receiver side, a variable optical attenuator (VOA) is applied to adjusted ROP before detecting by PD. A 20 GHz 3-dB bandwidth PIN with transimpedance amplifier (TIA) is utilized for O/E conversion. Then, the converted signal is captured by an Oscilloscope with 80-GSa/s sampling rate and 33 GHz bandwidth. For the receiver side DSPs, ANF is implemented to eliminate the clock leakage narrowband interference. And a RRC filter with roll off of 0.125 is used for match filtering. Then, signal synchronization and equalization with 47-taps FFE and 5-taps DFE are adopted for signal equalization. Finally, the BER is calculated after signal de-mapping. Due to the lack of high-bandwidth components in our hand, complicated DSP algorithms are necessary for coping with the ISI introduced by electronics components bandwidth limitation. When high bandwidth components are used in our experiment, the complicated DSP algorithms can be effectively avoided, and then the computation complexity of DSP algorithms in our experiment can be reduced significantly [32].
Fig. 3. Experimental setup. (a) Optical spectra (0.01 nm resolution) of 56Gbaud PAM-4 signal with and without pre-equalization
3.2 Effect of ANF
Due to the DAC clock leakage, the narrowband interference at ±20 GHz can be observed as shown in Fig. 4(a), which will seriously deteriorate the system performance. An ANF is implemented to suppress the narrowband interference at ±20 GHz. The effect of narrowband interference cancellation is verified by the 56 Gbaud PAM-4 signal with 4-bit resolution DAC and NS technique in OBTB. Figure 4 gives BER versus ROP of 56 Gbaud PAM-4 signal in OBTB with and without ANF. As shown in Fig. 4, a 1.5 dB receiver sensitivity improvement can be obtained at the HD-FEC threshold by utilizing ANF. Figure 4(b) gives the electrical spectra of the received 56 Gbaud PAM-4 signal with ANF. Compared with Fig. 4(a), we can find that the narrowband interference of signal at ±20 GHz is effectively eliminated. The eye diagrams of signal with ROP of −11dBm without and with ANF are shown in Fig. 4(c) and 4(d), respectively. It can be seen that, the eye diagram of received signal with ANF is clearer than the one without ANF.
Fig. 4. BER versus ROP of 56Gbaud PAM-4 in OBTB with and without ANF. Electrical spectra of (a) with and (b) without ANF of 56Gbaud PAM-4 signal in receiver side. Eye diagrams (c) without and (d) with ANF with ROP of −11dBm.
3.3 Transmission performance analysis with low resolution DAC
In order to verify the effectiveness of NS technique in low resolution DAC system, we firstly test the BER performance of signal over OBTB and 40-km SMF transmission. BER versus ROP of 50/56/60 Gbaud PAM-4 signal over OBTB and 40-km SMF transmission in O-band are given in Figs. 5(a), 5(b) and 5(c), respectively. The BER performance of 50 Gbaud and 56 Gbaud PAM-4 signal generated by 4-bit resolution DAC and NS technique can approximately approach that of signal generated by 8-bit resolution DAC both in OBTB and 40-km SMF transmission. And the receiver sensitivity can be improved by 1.1-dB and 2.8-dB at the BER of 3.8×10−3 when 4-bit resolution DAC is utilized, respectively. For 60 Gbaud PAM-4 signal, the BER is improved from 1.9×10−2 to 3.5×10−3 at ROP of −9dBm after OBTB and 40-km SMF transmission in O-band. Figures 6(a), 6(b) and 6(c) show eye diagrams of 56 Gbaud PAM-4 signal with 4-bit DAC without NS technique, 4-bit DAC with NS technique and 8-bit DAC at −12 dBm ROP, respectively. The eye opening becomes clear when NS technique is adopted in the transmission of PAM-4 signal generated with 4-bit resolution DAC, and no obvious improvement is achieved when 8-bit resolution DAC is applied for PAM-4 signal generation. According to the experimental results, the proposed 100 Gb/s/λ PAM-4 system with 4-bit resolution DAC and NS technique is a cost-efficient solution for 40-km Inter-DCI.
Fig. 5. BER versus ROP of (a) 50Gbaud PAM-4, (b) 56Gbaud PAM-4 and (c) 60Gbaud PAM-4 in OBTB and 40 km SMF transmission.
Fig. 6. Eye diagram of 56Gbaud PAM-4 signal with (a) 4-bit DAC w/o NS (b) 4-bit DAC w/ NS (c) 8-bit DAC over OBTB with ROP of −12dBm.
To satisfy 80-km SMF coverage in inter-DCI, 100 Gb/s/λ PAM-4 signal over 80-km SMF transmission with 4-bit resolution DAC in O-band is also investigated. A SOA and an OBPF are used to amplify the optical signal and minimize the out-of-band ASE noise before direct detection. However, ASE noise of SOA increases with the increase of current. Besides, the nonlinear impairments of TIA will degrade system performance at high received optical power. Thus, there is a trade-off between gains through signal amplification and penalties due to ASE and nonlinear impairments [21,22]. Figure 7(a) gives the BER versus SOA bias current of 50 Gbaud PAM-4 at received optical power of −20.4-dBm and −16-dBm before SOA. Then the optimized SOA bias current is set as 200 mA for 80-km transmission experiments. Figure 7(b) shows BER versus ROP of 50 Gbaud PAM-4 signal over 80-km SMF transmission in O-band. Due to the induced ASE noise of SOA, the BER performance of 50 Gbaud PAM-4 signal over 80-km transmission becomes worse at the same ROP compared with that over OBTB and 40-km SMF transmission shown in Fig. 7(b). Signal with 4-bit resolution DAC cannot reach the HD-FEC threshold, but can successfully reach HD-FEC after implementing NS technique at ROP of −6-dBm. Besides, similar to OBTB and 40-km SMF transmission, the BER performance of 50 Gbaud PAM-4 signal generated by 4-bit resolution DAC and NS technique can approach that of 50 Gbaud PAM-4 signal generated by 8-bit resolution DAC.
Fig. 7. (a) BER of 50Gbaud PAM-4 versus SOA bias current. (b) BER versus ROP for 50Gbaud PAM-4 over 80-km SMF transmission.
4. Conclusion
A low-cost 100 Gb/s/λ PAM-4 signal over 40/80-km SMF transmission in O-band utilizing 4-bit resolution DAC with BER under 3.8×10−3 is experimentally demonstrated in this paper. NS technique is used to reduce quantization noise of DAC, and the BER performance of 50 Gbaud PAM-4 signal generated by 4-bit resolution DAC and NS technique can approach that of 50 Gbaud PAM-4 signal generated by 8-bit resolution DAC over 40/80-km SMF transmission. Experiment results show that 100 Gb/s/λ PAM-4 signals over 40/80-km SMF transmission in O-band utilizing 4-bit resolution DAC is a promising solution for beyond 40-km Inter-DCI.
Funding
National Key R&D Program of China (2018YFB1800902); Local Innovation and Research Teams Project of Guangdong Pearl River Talents Program (2017BT01X121); Fundamental and Applied Basic Research Project of Guangzhou City under Grant (202002030326); Open Fund of IPOC (BUPT) (IPOC2020A010); National Natural Science Foundation of China (NSFC) (61871408).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. G. N. Liu, L. Zhang, T. Zuo, and Q. Zhang, “IM/DD Transmission Techniques for Emerging 5G Fronthaul, DCI, and Metro Applications,” J. Lightwave Technol. 36(2), 560–567 (2018). [CrossRef]
2. N. Eiselt, J. Wei, H. Griesser, A. Dochhan, M. Eiselt, J. Elbers, J. José, V. Olmos, and I. Tafur Monroy, “First Real-Time 400G PAM-4 Demonstration for Inter-Data Center Transmission over 100 km of SSMF at 1550 nm,” in Proc. Opt. Fiber Commun. Conf., 2016, paper W1 K.5.
3. J. Shi, J. Zhang, Y. Zhou, Y. Wang, and N. Chi, “Transmission Performance Comparison for 100-Gb/s PAM-4, CAP-16, and DFT-S OFDM With Direct Detection,” J. Lightwave Technol. 35(23), 5127–5133 (2017). [CrossRef]
4. W. Wang, F. Li, Z. Li, Q. Sui, and Z. Li, “Dual drive Mach-Zehnder Modulator-based single side-band modulation direct detection system without signal-to-signal beating interference,” J. Lightwave Technol. 38(16), 4341–4351 (2020). [CrossRef]
5. 802.3cu-2021 - IEEE Standard for Ethernet. 2021. [Online]. Available: https://standards.ieee.org/standard/802_3cu-2021.html. Accessed Feb. 9, 2021.
6. R. Matsumoto, K. Matsuda, and N. Suzuki, “Fast, Low-Complexity Widely-Linear Compensation for IQ Imbalance in Burst-Mode 100-Gb/s/λ Coherent TDM-PON,” in Proc. Opt. Fiber Commun. Conf., 2018, paper M3B.2.
7. Y. Zhu, M. Jiang, X. Ruan, C. Li, and F. Zhang, “16×112Gb/s Single-Sideband PAM4 WDM Transmission over 80 km SSMF with Kramers-Kronig Receiver,” in Proc. Opt. Fiber Commun. Conf., 2018, paper Tu2D.2.
8. H. Li, M. Luo, X. Li, and S. Yu, “Demonstration of 50-Gb/s/λ PAM-4 PON with Single-PD using Polarization-Insensitive and SSBI Suppressed Heterodyne Coherent Detection,” in Proc. Opt. Fiber Commun. Conf., 2020, paper
9. M. Sezer Erkilinc, S. Pachnicke, H. Griesser, B. C. Thomsen, P. Bayvel, and R. I. Killey, “Dispersion-precompensated direct-detection Nyquist-pulse-shaped subcarrier modulation using a dual-drive Mach–Zehnder modulator,” in OECC, 2015, paper 1570087413.
10. D. Lai, Z. Li, D. Zou, X. Yi, Z. Li, and F. Li, “Transmission of 50-Gb/s/λ PAM-4 over 100-km SMF for LR-PON utilizing Low Complexity Polarization-insensitive Quasi-coherent Receiver,” in ACP, 2020, paper M4A.289.
11. M. Yin, D. Zou, X. Yi, Z. Li, and F. Li, “Low Power-Consumption 50-Gb/s/λ PON Utilizing BPAM-4 Modulation,” in ACP, 2020, paper M4A.225.
12. D. Zou, F. Li, Z. Li, W. Wang, Q. Sui, Z. Cao, and Z. Li, “100G PAM-6 and PAM-8 signal transmission enabled by pre-chirping for 10-km intra-DCI utilizing MZM in C-band,” J. Lightwave Technol. 38(13), 3445–3453 (2020). [CrossRef]
13. S. Yamamoto, N. Edagawa, H. Taga, Y. Yoshida, and H. Wakabayashi, “Analysis of laser phase noise to intensity noise conversion by chromatic dispersion in intensity modulation and direct detection optical-fiber transmission,” J. Lightwave Technol. 8(11), 1716–1722 (1990). [CrossRef]
14. A. R. Charaplyvy, R. W. Tkach, L. L. Buhl, and R. C. Alferness, “Phase modulation to amplitude modulation conversion of CW laser light in optical fibres,” Electron. Lett. 22(8), 409–411 (1986). [CrossRef]
15. F. Devaux, Y. Sorel, and J. Kerdiles, “Simple measurement of fiber dispersion and of chirp parameter of intensity modulated light emitter,” J. Lightwave Technol. 11(12), 1937–1940 (1993). [CrossRef]
16. D. Zou, Y. Chen, Z. Li, F. Li, L. Ding, Y. Sun, J. Li, Q. Sui, X. Yi, and Z. Li, “Comparison of null-subcarriers reservation and adaptive notch filter for narrowband interference cancellation in intra-data center interconnect with DMT signal transmission,” Opt. Express 27(4), 5696–5702 (2019). [CrossRef]
17. F. Li, D. Zou, Q. Sui, J. Li, X. Yi, L. Li, and Z. Li, “Optical Amplifier-free 100 Gbit/s/lamda PAM-N Transmission and Reception in O-band over 40-km SMF with 10-G Class DML,” in Proc. Opt. Fiber Commun. Conf., 2019, paper Tu2F.4.
18. D. Zou, F. Li, W. Wang, Z. Li, and Z. Li, “Amplifier-less transmission of beyond 100-Gbit/s/λ signal for 40-km DCI-Edge with 10G-class O-band DML,” J. Lightwave Technol. 38(20), 5649–5655 (2020). [CrossRef]
19. K. Zhong, X. Zhou, J. Huo, H. Zhang, J. Yuan, Y. Yang, C. Yu, A. Lau, and C. Lu, “Amplifier-less transmission of single channel 112 Gbit/s PAM4 signal over 40 km using 25G EML and APD at O band,” in 2017 European Conference on Optical Communication (ECOC), 2016, pp. 1–3.
20. W. Wang, P. Zhao, Z. Zhang, H. Li, D. Zang, N. Zhu, and Y. Lu, “First Demonstration of 112 Gb/s PAM-4 Amplifier-free Transmission over a Record Reach of 40 km Using 1.3 µm Directly Modulated Laser,” in Proc. Opt. Fiber Commun. Conf., 2018, paper Th4B.8.
21. J. Zhang, J. Yu, H. Chien, J. S. Wey, M. Kong, X. Xin, and Y. Zhang, “Demonstration of 100-Gb/s/λ PAM-4 TDM-PON Supporting 29-dB Power Budget with 50-km Reach Using 10G-class O-band DML Transmitters,” in Proc. Opt. Fiber Commun. Conf., 2019, paper Th4C.3.
22. J. Zhang, J. Yu, J. S. Wey, X. Li, L. Zhao, K. Wang, M. Kong, W. Zhou, J. Xiao, X. Xin, and F. Zhao, “SOA Pre-Amplified 100 Gb/s/λ PAM-4 TDM-PON Downstream Transmission Using 10 Gbps O-Band Transmitters,” J. Lightwave Technol. 38(2), 185–193 (2020). [CrossRef]
23. K. Wang, J. Zhang, Y. Wei, L. Zhao, W. Zhou, M. Zhao, J. Xiao, X. Pan, B. Liu, X. Xin, L. Zhang, Y. Zhang, and J. Yu, “100-Gbit/s/λ PAM-4 signal transmission over 80-km SSMF based on an 18-GHz EML at O-band,” in Proc. Opt. Fiber Commun. Conf., 2020, paper Th1D.5.
24. W. A. Ling, “Shaping Quantization Noise and Clipping Distortion in Direct-Detection Discrete Multitone,” J. Lightwave Technol. 32(9), 1750–1758 (2014). [CrossRef]
25. M. Yin, W. Wang, D. Zou, Q. Sui, X. Yi, Z. Li, and F. Li, ““Low Cost 100Gb/s/λ PAM-4 Signal Transmission for 40-km Inter-DCI with 4-bit Resolution DAC in O-band,” in Proc. Opt. Fiber Commun. Conf., 2021, accept, control number: 3560650.
26. H. Li, R. Hu, Q. Yang, M. Luo, Z. He, P. Jiang, Y. Liu, X. Li, and S. Yu, “Improving performance of mobile fronthaul architecture employing high order delta-sigma modulator with PAM-4 format,” Opt. Express 25(1), 1–9 (2017). [CrossRef]
27. J. Wang, Z. Jia, L. A. Campos, C. Knittle, and G. Chang, “Optical Coherent Transmission of 20 × 192-MHz DOCSIS 3.1 Channels with 16384QAM based on Delta-Sigma Digitization,” in Proc. Opt. Fiber Commun. Conf., 2017, paper Th1 K.1.
28. Y. Yoffe, G. Khanna, E. Wohlgemuth, E. d. Man, B. Spinnler, N. Hanik, A. Napoli, and D. Sadot, “Low-Resolution Digital Pre-Compensation Enabled by Digital Resolution Enhancer,” J. Lightwave Technol. 37(6), 1543–1551 (2019). [CrossRef]
29. T. Tanaka, M. Nishihara, T. Takahara, W. Yan, L. Li, Z. Tao, M. Matsuda, K. Takabayashi, and J. C. Rasmussen, “Experimental Demonstration of 448-Gbps+ DMT Transmission ove r 30-km SMF,” in Proc. Opt. Fiber Commun. Conf., 2014, paper M2I.5.
30. F. Li, D. Zou, L. Ding, Y. Sun, J. Li, Q. Sui, L. Li, X. Yi, and Z. Li, “100 Gbit/s PAM4 signal transmission and reception for 2-km interconnect with adaptive notch filter for narrowband interference,” Opt. Express 26(18), 24066–24074 (2018). [CrossRef]
31. X. Pang, J. Van Kerrebrouck, O. Ozolins, R. Lin, A. Udalcovs, L. Zhang, S. Spiga, M. C. Amann, G. Van Steenberge, L. Gan, M. Tang, S. Fu, R. Schatz, G. Jacobsen, S. Popov, D. Liu, W. Tong, G. Torfs, J. Bauwelinck, X. Yin, and J. Chen, “7×100 Gbps PAM-4 Transmission over 1-km and 10-km Single Mode 7-core Fiber using 1.5-µm SM-VCSEL,” in Proc. Opt. Fiber Commun. Conf., 2018, paper M1I.4.
32. N. P. Diamantopoulos, W. Kobayashi, H. Nishi, K. Takeda, T. Kakitsuka, and S. Matsuo, “Amplifierless PAM-4/PAM-8 transmissions in O-band using a directly modulated laser for optical data-center interconnects,” Opt. Lett. 44(1), 9–12 (2019). [CrossRef]
