4 &#x00D7; 128-Gb/s PDM-DMT signal transmission over 1440-km SSMF with high phase noise tolerance

Cai Li; Ming Luo; Xiang Li

doi:10.1364/OE.26.030901

1. Introduction

Intensity modulation and direct detection (IM-DD) system has been widely employed in the short and medium reach optical transmission system due to the characteristics of simple implementation and low cost [1–3]. The continuing traffic growth in metropolitan area networks has created a dire need for high speed transmission and extendible transmission distance. However, IM-DD system suffers from severe power fading induced by the chromatic dispersion (CD), resulting in a limited transmission distance [4]. The carrier-assisted single side band modulation with direct detection (SSB-DD) scheme has been proposed to overcome the power fading problem [5–9]. However, such scheme suffers from severe signal-to-signal beating noise (SSBN) which degrades the system performance. Several schemes are proposed to reduce the SSBN, such as SSBN cancellation scheme [10–12] and Kramers-Kronig (KK) field reconstruction [13]. However, these systems are usually based on complex IQ modulation and unable to achieve polarization division multiplexing (PDM) without more complex receiver structure or digital signal processing (DSP) due to the random assisted carrier splitting at the received side. The 4 × 4 multi-input-multi-output (MIMO) technique with two orthogonal polarized assisted carriers is employed to realize the PDM-SSB-DD system [14]. Another approach to achieve direct detection of PDM signal is based on Stokes receiver [15–17], which requires three photodiodes (PDs) as well as three analogue-to-digital converters (ADCs) to detect the PDM signal.

On the other hand, the low-cost laser with high line-width can be used in traditional IM-DD system because the phase noise has no effects on the system performance. However, phase noise compensation is generally required in carrier-assisted SSB-DD system due to the phase difference between the assisted carrier and the SSB signal. Although in some proposed schemes [10], the optical signal and assisted carrier share the same laser source, the corresponding phases may be different after long-distance transmission. Therefore, based on conventional phase noise compensation technique, narrow line-width lasers are used as the modulated carrier and the assisted carrier. In our previously work, digital carrier regeneration (DCR) is proposed to achieve a high phase noise tolerance [18]. Different from conventional intensity modulation, the intensity signal is modulated onto a Mach-Zehnder modulator (MZM) by setting the bias around the null point, resulting in a suppressed optical carrier transmission with the intensity signal. At the receiver side, the intensity signal can be recovered by envelope detection and the DCR method. Although the spectrum efficiency of intensity modulation is half of the complex modulation, the laser linewidth tolerance is more robust compared to the complex modulation scheme [19].

In this paper, intensity modulation and coherent detection (IM-CohD) scheme is employed to balance the low cost requirement and system performance. At the transmitter side, discrete multi-tone (DMT) signal is modulated by an MZM and transmitted with a suppressed optical carrier. A local oscillator (LO) is used at the receiver side to amplify the received signal by heterodyne detection. A pair of polarization beam splitters (PBSs) are employed for PDM signal detection. By applying the DCR method, the coherent transceiver using directly modulated lasers (DMLs) at both transmitter and receiver sides is successfully demonstrated with high line-width tolerance. In the experiment, 4 × 128-Gb/s KK and DCR based PDM-DMT signal is transmitted over 1440-km SSMF with DMLs as optical carrier and LO. Experiment results show that KK and DCR scheme can mutually improve system performance.

2. Principal of KK and DCR based receiver

To illustrate the principle of KK and DCR based receiver in PDM-DMT system, we first introduce the DSP of single polarization DMT signal for simplicity, as shown in Fig. 1. To satisfy the minimum phase condition of KK detection technique, the LO is located at the edge of the DMT signal, where the frequency offset $Δ f$ should be no less than half of the bandwidth of DMT signal. After optical-to-electrical conversion and analog-to-digital conversion, the digital received signal can be expressed as

\begin{matrix} r (n) = {| A_{l o} + (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))} |}^{2} \\ = {| A_{l o} |}^{2} + {| A_{c} + s (n) |}^{2} \\ + A_{l o} \cdot (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))} + A_{l o} \cdot (A_{c} + s (n)) e^{- j (2 π Δ f n + Δ φ (n))} \end{matrix}

where

s (n)

is the transmitted DMT signal,

A_{l o}

and

A_{c}

is the constant amplitude of LO and the suppressed transmitted carrier,

Δ φ (n)

represents the phase difference between the suppressed transmitted carrier and LO at the n_th sampling point. The received electrical signal is consisted of four components:

{| A_{l o} |}^{2}

is the beating of LO,

{| A_{c} + s (n) |}^{2}

is the SSBN,

A_{l o} \cdot (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))}

and

A_{l o} \cdot (A_{c} + s (n)) e^{- j (2 π Δ f n + Δ φ (n))}

denote the right sideband signal and the left sideband signal. As the frequency offset

Δ f

is no less than half of the bandwidth of DMT signal, there is no frequency overlap between the right sideband signal and the left sideband signal as shown in Fig. 1(b), where the conventional CD compensation can be employed in our receiver DSP to mitigate the influence of chromatic dispersion with or without KK detection.

Fig. 1 Schematic diagram of receiver DSP. (a) Optical spectrum of received signal; (b) electrical spectrum of received signal; (c) electrical spectrum after KK Scheme; (d) electrical spectrum after frequency offset compensation.

Download Full Size | PDF

To improve the receiver performance, KK detection technique is employed to reduce the SSBN. As shown in Fig. 2(a), the amplitude $A_{k k} (n)$ and the phase $θ_{k k} (n)$ of the KK recovered signal $r_{k k} (n)$ can be calculated as

\begin{matrix} A_{k k} (n) = \sqrt{r (n)} = | A_{l o} + (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))} | \\ θ_{k k} (n) = H (\ln (A_{k k} (n))) \end{matrix}

where

H (\cdot)

and

\ln (\cdot)

denote the Hilbert transform and logarithm operation, respectively. Then, the signal after KK recovery can be expressed as

\begin{matrix} r_{k k} (n) = | A_{l o} + (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))} | \cdot e^{j θ_{k k} (n)} \\ = (A_{l o} + (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))}) \cdot e^{- j θ_{s i g} (n)} \cdot e^{j θ_{k k} (n)} \\ = (A_{l o} + (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))}) \cdot e^{j Δ θ (n)} \end{matrix}

where

θ_{s i g} (n)

represents the true phase of the signal

A_{l o} + (A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n))}

and

Δ θ (n)

represents the residual phase noise between

θ_{s i g} (n)

and

θ_{k k} (n)

.

Fig. 2 (a) Schematic diagram of KK detection; (b) Schematic diagram of the DCR scheme; (c) Recovered DMT signal with regenerated digital carrier.

Download Full Size | PDF

After CD compensation and DC removing, the signal can be expressed as $(A_{c} + s (n)) e^{j (2 π Δ f n + Δ φ (n) + Δ θ (n))}$ . The frequency offset $2 π Δ f n$ can be easily compensated by the conventional frequency offset compensation method. For DMT signal $s (n)$ , if the suppressed transmitted carrier $A_{c}$ is large enough to ensure $A_{c} + s (n) > 0$ , we could directly remove the residual phase by modulo operation. However, the transmitted carrier is suppressed to achieve high optical power efficiency, where the value of $A_{c} + s (n)$ is not always larger than zero. Hence, the DCR algorithm is employed after frequency offset compensation to mitigate the phase noise $Δ φ (n)$ caused by laser linewidth as well as $Δ θ (n)$ caused by KK detection. The signal after frequency offset compensation can be expressed as

\begin{matrix} r_{D C R_i n} (n) = (A_{c} + s (n)) e^{j (Δ φ (n) + Δ θ (n))} \\ = A_{c} \cdot e^{j (Δ φ (n) + Δ θ (n))} + s (n) \cdot e^{j (Δ φ (n) + Δ θ (n))} \end{matrix}

As shown in Fig. 2(b), the DCR scheme firstly extracts the suppressed transmitted carrier $c (n) = A_{c} \cdot e^{j (Δ φ (n) + Δ θ (n))}$ from $r_{D C R_i n} (n)$ by a narrow digital low pass filter (LPF). Then, the suppressed transmitted carrier $c (n)$ is amplified by a factor of $α$ , obtaining an amplified digital carrier $α \cdot c (n)$ . Finally, the recovered DMT signal can be obtained by

\begin{matrix} r_{D C R_o u t} (n) = | r_{D C R_i n} (n) + α \cdot c (n) | - | α \cdot c (n) | \\ = | (α + 1) \cdot A_{c} + s (n) | - | α \cdot A_{c} | \\ = A_{c} + s (n), α ≫ 1 \end{matrix}

By choosing appropriate amplifying factor

α

, we can assume that

(α + 1) \cdot A_{c} + s (n) > 0

. The time domain signal

r_{D C R_o u t} (n)

with

α = 100

is shown in Fig. 2(c), as well as the regenerated digital carrier

α \cdot c (n)

shown by the red line. Since the line-width of DML is large, the amplitude of regenerated digital carrier suffers from severe fluctuations. After DCR, we can recover the DMT signal by envelop detection without phase noise compensation. Then, the conventional DSP operations for DMT signal are employed to recover the original data, including time synchronization, fast Fourier transform (FFT), channel estimation, and symbol decision.

3. Experimental setup

Figure 3 shows the experimental setup for the transmission of 4 × 128-Gb/s PDM-DMT signal over 1440-km SSMF. At the transmitter, four laser sources are combined by a polarization maintaining optical coupler (PMOC) and then split into two polarizations by a polarization beam splitter (PBS). One of the four laser source is a DML with linewidth of ~10MHz. The DMT signals in two polarizations are generated by an arbitrary waveform generator (AWG) operating at 56-GSa/s. The FFT size is 256, where the first 4 subcarriers are used as the guard band. The following 76 subcarriers are loaded with data in 16- quadrature amplitude modulation (QAM) format. Hermitian symmetry is then used to produce real-valued output. The frequency gap for the low pass filter in DCR scheme is ~1.75 GHz, where ~5% of the spectral efficiency (4 subcarriers of totally 80 subcarriers) is sacrificed. 1/32 of the FFT size is used as cyclic prefix (CP) to avoid the inter-symbol interference. The generated DMT signals from the two outputs of the AWG are then modulated onto the two MZMs, which are biased around the null point for power efficient transmission. The modulated optical signals are then combined by a polarization beam combiner (PBC) to form the PDM-DMT signal with net bit rate of 128.97-Gb/s per channel. Then, each band PDM-DMT signal is delayed by several pieces of fiber for decorrelation. The generated multi-band optical spectrum at the transmitter side is shown in Fig. 3(a). The transmission link is constructed by 24 spans of 80-km SSMF with Raman amplifier only. The spectrum of the multi-band signal after 1440-km transmission is shown in Fig. 3(b). At the receiver side, optical tunable filter (OTF) is used to selected the desired subband and an Erbium-doped optical fiber amplifier (EDFA) is used to adjust the received optical power. The single tone PDM-DMT spectrum is shown in Fig. 3(c). Then, the signal is first split by a PBS and each polarization is combined with an LO tapped from another DML. The LO’s wavelength is located at the edge of the optical DMT signal. The spectrum of the received optical signals combined with LO is shown in Fig. 3(d). After optical-to-electrical conversion by two PDs, the two electrical signals are then fed to a LeCroy real-time scope, acquired at sampling rate of 160-GSa/s, and processed offline DSP. In the offline processing, we investigate four DSP cases for comparison, those are, DSP-1: the KK and DCR based PDM-DMT system, which is illustrated in Section 2; DSP-2: the DCR based PDM-DMT system, where the SSBN is not mitigated; DSP-3: the KK based PDM-DMT system which employs conventional phase noise (PN) compensation scheme; DSP-4: the conventional DSP-based PDM-DMT system. To evaluate the tolerance of laser phase noise based on the proposed method, the performance based on external cavity lasers (ECLs) with linewidth of ~50-kHz as LO and optical carrier is also investigated for fair comparison.

Fig. 3 Experimental setup and four DSP cases for the transmission of 4 × 100-Gb/s PDM-DMT signal over 1440-km SSMF. (a) 4 × 128-Gb/s modulated PDM-DMT spectrum at the transmitter; (b) 4 × 128-Gb/s modulated PDM-DMT spectrum after 1440km transmission; (c) filtered 128-Gb/s modulated PDM-DMT spectrum at the receiver; (d) the spectrum of received optical signals combined with LO.

Download Full Size | PDF

4. Results and discussion

Firstly, we investigate the carrier-to-signal power ratio (CSPR) versus bit error rate (BER) of the proposed system at back to back case. The ECL-based coherent detection is employed for comparison, where two ECLs are employed as the modulated optical carrier and LO at the transmitter and receiver sides. As shown in Fig. 4(a), the performance comparison between DSP-1 and DSP-2 shows that the BER degradation caused by the SSBN is serious with DSP-2 when CSPR is low, and KK recovery technique can provide a dramatic improvement with DSP-1. The same situation can be found in the comparison between DSP-3 and DSP-4, where KK technique can improve the BER performance at low CSPR. While CSPR is larger than 23-dB, the improvement by KK technique with DSP-1 is negligible compared to using DSP-2 as well as DSP-3 compared with DSP-4. The reason is that a high CSPR can reduce the influence of SSBN, and the BER performance is mainly limited by the OSNR. The optimal CSPR is ~16-dB for DSP-1 and ~23-dB for DSP-2. The performance difference between DSP-1 and DSP-3 shows that the phase noise caused by KK detection cannot be compensated by the conventional phase compensation scheme while DCR can reduce the effects of phase noise. The optimal CSPR is ~16-dB for DSP-3 which is in accordance with DSP-1, and it is ~19-dB for DSP-4. It is worth mentioning that the optimal CSPR in the proposed method is a bit larger than that of other reported KK detection method. This is mainly because the SSBN could be larger than that in other reported method, where the SSBI is illustrated as ${| A_{c} + s (n) |}^{2}$ instead of ${| s (n) |}^{2}$ according to Eq. (1). With the corresponding optimal CSPR, we also investigate the OSNR versus BER performance in back to back case with the four DSPs. As shown in Fig. 4(b), the performance of DSP-1 and DSP-2 with low OSNR is almost the same due to channel noise is the major limiting factor. As the OSNR increasing, the SSBN becomes the dominant limiting factor, and DSP-1 shows better BER performance with SSBN mitigation by KK technique than DSP-2. The required OSNR at the BER threshold of 2.0 × 10⁻² is found to be 21.8-dB, 22.4-dB, 26.1-dB, and 28.4-dB for DSP-1, DSP-2, DSP-3, and DSP-4, respectively. The OSNR improvement provided by KK technique is 0.6-dB comparing with DSP-1 and DSP-2, and 2.3-dB comparing with DSP-3 and DSP-4, which also shows that channel noise is the major limiting factor at low OSNR. It is also shown that the DCR scheme can provide 4.3-dB OSNR improvement comparing with DSP-1 and DSP-3, and 6.0-dB OSNR improvement comparing with DSP-2 and DSP-4, which indicate that the DCR scheme has the potential to mitigate high laser phase noise.

Fig. 4 (a) CSPR versus BER with ECLs at back to back case; (b) OSNR versus BER with ECLs at back to back case.

Download Full Size | PDF

Then, the CSPR versus BER of the four cases using DMLs as LO and modulated optical carrier are also investigated in the back to back case, as shown in Fig. 5(a). It is obviously that, due to the large linewidth of DML, the received signal cannot be recovered with DSP-3 and DSP-4, which means DCR is an important scheme for the phase noise compensation in the coherent transceiver structure using high line-width lasers. The performance comparison between DSP-1 and DSP-2 shows similar trends as those in the transceiver using ECLs. The optimal CSPR is ~16-dB for DSP-1 and ~23-dB for DSP-2 in Fig. 5(a), which are the same as those in Fig. 4(a). Figure 5(b) shows the OSNR versus BER with the four DSPs in transceiver using DMLs with the corresponding optimal CSPR. Similar to the back-to-back case, the received signal cannot be recovered with DSP-3 and DSP-4. The required OSNR at the BER threshold of 2.0 × 10⁻² is found to be 22.4 dB and 23.7 dB with DSP-1 and DSP-2, respectively, which shows KK recovery technique provide additional 1.3 dB OSNR improvement. Moreover, compared with the case using ECLs as LO and optical carrier, there is only 0.6 dB OSNR penalty caused by large linewidth of DML after phase noise compensation by the DCR scheme.

Fig. 5 (a) CSPR versus BER with DMLs at back to back case; (b) OSNR versus BER with DMLs at back to back case.

Download Full Size | PDF

We also investigate the BER performance versus alpha factor in DCR scheme with the optimum CSPR value, as shown in Fig. 6. The BER performance is convergent with alpha factor larger than 100 in the ECL-based system, and the required alpha factor should be larger than 500 while using DMLs. However, the performance would not be improved or degraded by enlarging the alpha factor, when the amplified factor is larger than a certain value.

Fig. 6 BER performance versus alpha factor of DCR scheme at back-to-back case.

Download Full Size | PDF

Figure 7(a) shows launch power versus BER performance for the transmission of 4 × 128-Gb/s PDM-DMT signal in transceiver using ECLs over 1040-km SSMF. The optimal launch power is 2-dBm for all the four DSPs. Under the optimal launch power, the transmission distance versus BER performance is investigated, as shown in Fig. 7(b). The signal with DSP-4 can reach ~1200-km transmission at the BER threshold of 2.0 × 10⁻². The transmission distance can be improved to more than 1500-km by employing KK recovery technique in DSP-2. With the DCR method in DSP-1, the transmission distance can be enhanced to more than 2000-km with a large margin.

Fig. 7 (a) Launch power versus BER performance for the transmission of 4 × 128-Gb/s PDM-DMT signal using ECLs over 1040-km SSMF; (b) transmission distance versus BER performance with ECLs of 4 × 128-Gb/s PDM-DMT signal.

Download Full Size | PDF

The launch power versus BER performance for the transmission of 4 × 128-Gb/s PDM-DMT signal using DMLs over 1040-km SSMF is investigated, as shown in Fig. 8(a). As the same as that in back to back case, the received signal cannot be recovered with DSP-3 and DSP-4. The optimal BER is 9.0 × 10⁻³ and 1.9 × 10⁻² with launch power ~3-dBm for DSP-1 and DSP-2, respectively. Figure 8(b) shows the transmission distance versus BER performance using DMLs with the optimal launch power of 3-dBm. Under the 20% FEC threshold, the 4 × 128-Gb/s PDM-DMT signal can be transmitted over 1200-km via the DCR method. Moreover, the signal after KK recovery and DCR techniques can reach ~1440-km SSMF transmission. It is also shown that the performance degradation is larger when DMLs are used as optical carrier and LO. This is mainly because the linewidth of DMLs could be further broadening after fiber transmission, which results additional performance penalty.

Fig. 8 (a) Launch power versus BER performance with DMLs for the transmission of 4 × 128-Gb/s PDM-DMT signal over 1040-km SSMF; (b) transmission distance versus BER performance with DMLs of 4 × 128-Gb/s PDM-DMT signal.

Download Full Size | PDF

The comparison of hardware complexity is shown in Table 1. Comparing with the typical coherent detection (CohD) system, the proposed KK and DCR based IM-CohD system employs two single-ended PDs instead of four pairs of balanced PDs, and two digital-to-analog converters (DACs) and two analog-to-digital converters (ADCs) instead of four DACs and four DACs, respectively. However, the hardware complexity of the proposed scheme is almost twice of the typical DD system, since the proposed scheme employs polarization multiplexing technique. Table 2 shows the DSP complexity comparison of the proposed scheme, typical DD, and typical CohD systems. The typical DD system without DSP can achieve a transmission distance less than 40 km, and with the CD compensation, the distance can be extended up to 100 km. The maximum transmission distance of the proposed scheme can be extended to more than 1000 km based on the proposed DSP method, where the DSP complexity is almost the same as that of the typical CohD system. In conclusion, the proposed KK and DCR based IM-CohD scheme can provide a tradeoff between the low cost requirement and system performance.

Table 1. The hardware complexity comparisons of the proposed scheme, typical DD, and typical CohD systems.

View Table | View all tables in this article

Table 2. The DSP complexity comparisons of the proposed scheme, typical DD, and typical CohD systems.

View Table | View all tables in this article

5. Conclusion

In this paper, we propose a low-cost IM-CohD scheme for PDM-DMT signals with high phase noise tolerance based on DCR and KK detection. The KK detection can mitigate the SSBN, and the DCR method can eliminate both the phase noise caused by high line-width of laser and the residual phase noise after KK recovery. In the experiment, we investigate the performance improvement by KK recovery and DCR techniques in coherent optical transceiver structures employing ECLs and DMLs, respectively. It is shown that KK recovery and DCR techniques can mutually improve system performance. Moreover, the signal with large phase noise cannot be recovered without DCR due to the large laser line-width of DMLs. Finally, we successfully demonstrate 4 × 128-Gb/s PDM-DMT signal transmission over 1440-km SSMF using DMLs as optical carrier and LO based on KK recovery and DCR techniques, which has the potential applications for future cost-effective metro networks.

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. K. Zhong, X. Zhou, J. Huo, C. Yu, C. Lu, and A. P. T. Lau, “Digital signal processing for short-reach optical communications: a review of current technologies and future trends,” J. Lightwave Technol. 36(2), 377–400 (2018). [CrossRef]

3. J. Shi, J. Zhang, Y. Zhou, Y. Wang, N. Chi, and J. Yu, “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. D. Che, Q. Hu, and W. Shieh, “Linearization of direct detection optical channels using self-coherent subsystems,” J. Lightwave Technol. 34(2), 516–524 (2016). [CrossRef]

5. Y. Zhu, K. Zou, Z. Chen, and F. Zhang, “224Gb/s optical carrier-assisted Nyquist 16-QAM half-cycle singlesideband direct detection transmission over 160km SSMF,” J. Lightwave Technol. 35(9), 1557–1565 (2017). [CrossRef]

6. X. Chen, C. Antonelli, S. Chandrasekhar, G. Raybon, J. Sinsky, A. Mecozzi, M. Shtaif, and P. Winzer, “218- Gb/s single-wavelength, single-polarization, single-photodiode transmission over 125-km of standard single mode fiber using Kramers-Kronig detection,” Proc. Optical Fiber Communication Conference, Paper Th5B.6 (2017).

7. W. Peng, X. Wu, V. R. Arbab, K. Feng, B. Shamee, L. C. Christen, J. Yang, and A. E. Willner, “Theoretical and experimental investigations of direct-detected RF-tone assisted optical OFDM systems,” J. Lightwave Technol. 27(10), 1332–1339 (2009). [CrossRef]

8. T. M. F. Alves, L. M. M. Mendes, and A. V. T. Cartaxo, “High granularity multiband OFDM virtual carrier assisted direct-detection metro networks,” J. Lightwave Technol. 33(1), 42–54 (2015). [CrossRef]

9. S. T. Le, K. Schuh, M. Chagnon, F. Buchali, R. Dischler, V. Aref, H. Buelow, and K. M. Engenhardt, “1.72Tb/s virtual-carrier assisted direct-detection transmission over 200km,” J. Lightwave Technol. 36(6), 1347–1353 (2018). [CrossRef]

10. W. R. Peng, X. Wu, K. M. Feng, V. R. Arbab, B. Shamee, J. Y. Yang, L. C. Christen, A. E. Willner, and S. Chi, “Spectrally efficient direct-detected OFDM transmission employing an iterative estimation and cancellation technique,” Opt. Express 17(11), 9099–9111 (2009). [CrossRef] [PubMed]

11. K. Zou, Y. Zhu, F. Zhang, and Z. Chen, “Spectrally efficient terabit optical transmission with Nyquist 64-QAM half-cycle subcarrier modulation and direct detection,” Opt. Lett. 41(12), 2767–2770 (2016). [CrossRef] [PubMed]

12. Z. Li, M. Erkılınç, K. Shi, E. Sillekens, L. Galdino, B. Thomsen, P. Bayvel, and R. Killey, “SSBI mitigation and the Kramers-Kronig scheme in single-sideband direct-detection transmission with receiver-based electronic dispersion compensation,” J. Lightwave Technol. 35(10), 1887–1893 (2017). [CrossRef]

13. A. Mecozzi, C. Antonelli, and M. Shtaif, “Kramers-kronig coherent receiver,” Optica 3(11), 1220–1227 (2016). [CrossRef]

14. D. Qian, N. Cvijetic, J. Hu, and T. Wang, “108 Gb/s OFDMA-PON with polarization multiplexing and direct detection,” J. Lightwave Technol. 28(4), 484–493 (2010). [CrossRef]

15. T. M. Hoang, M. Y. S. Sowailem, Q. Zhuge, Z. Xing, M. Morsy-Osman, E. El-Fiky, S. Fan, M. Xiang, and D. V. Plant, “Single wavelength 480 Gb/s direct detection over 80km SSMF enabled by Stokes vector Kramers Kronig transceiver,” Opt. Express 25(26), 33534–33542 (2017). [CrossRef]

16. D. Che, C. Sun, and W. Shieh, “Single-channel 480-Gb/s direct detection of POL-MUX IQ signal using single side band Stokes vector receiver,” Proc. Optical Fiber Communication Conference, Tu2C.7 (2018).

17. C. Antonelli, A. Mecozzi, M. Shtaif, X. Chen, S. Chandrasekhar, and P. J. Winzer, “Polarization multiplexing with the Kramers-Kronig receiver,” J. Lightwave Technol. 35(24), 5418–5424 (2017). [CrossRef]

18. C. Li, R. Hu, Q. Yang, M. Luo, W. Li, and S. Yu, “Fading-free transmission of 124-Gb/s PDM-DMT signal over 100-km SSMF using digital carrier regeneration,” Opt. Express 24(2), 817–824 (2016). [CrossRef] [PubMed]

19. X. Chen, C. Antonelli, S. Chandrasekhar, G. Raybon, A. Mecozzi, M. Shtaif, and P. J. Winzer, “4 x 240 Gb/s Dense WDM and PDM Kramers-Kronig Detection with 125-km SSMF Transmission,” Proc. European Conference on Optical Communication, W.2.D.4 (2017).

	Laser linewidth	Modulator	Receiver	DAC	ADC
Proposed scheme	~10 MHz	DP-MZM	2 PDs	2	2
Typical DD	> 10 MHz	MZM	1 PD	1	1
Typical CohD	<100 KHz	DP-IQM	4 BPDs	4	4

	DSP	CD Comp.	FO Comp.	PN Comp.	Others	Max reach dist.
Proposed scheme	YES	YES	YES	DCR	KK	~1000 km
Typical DD	YES	YES	NO	NO	NO	<100 km
Typical DD	NO	NO	NO	NO	NO	<40 km
Typical CohD	YES	YES	YES	YES	IQ skew Comp.	>2000 km

	Laser linewidth	Modulator	Receiver	DAC	ADC
Proposed scheme	~10 MHz	DP-MZM	2 PDs	2	2
Typical DD	> 10 MHz	MZM	1 PD	1	1
Typical CohD	<100 KHz	DP-IQM	4 BPDs	4	4

	DSP	CD Comp.	FO Comp.	PN Comp.	Others	Max reach dist.
Proposed scheme	YES	YES	YES	DCR	KK	~1000 km
Typical DD	YES	YES	NO	NO	NO	<100 km
Typical DD	NO	NO	NO	NO	NO	<40 km
Typical CohD	YES	YES	YES	YES	IQ skew Comp.	>2000 km

4 × 128-Gb/s PDM-DMT signal transmission over 1440-km SSMF with high phase noise tolerance

Abstract

1. Introduction

2. Principal of KK and DCR based receiver

3. Experimental setup

4. Results and discussion

5. Conclusion

References

Cited By

Figures (8)

Tables (2)

Equations (5)

Optics Express