We investigated through simulations the performance of Nyquist-WDM Terabit superchannels implemented using polarization-multiplexed phase shift-keying based on 2 (PM-BPSK) and 4 (PM-QPSK) signal points or polarization-multiplexed quadrature amplitude modulation based on 8 (PM-8QAM) and 16 (PM-16QAM) signal points. Terabit superchannels are obtained through the aggregation of multiple subcarriers using the Nyquist-WDM technique, based on a tight spectral shaping of each subcarrier which allows very narrow spacing. We first studied the optimum transmitter/receiver filtering in a back-to-back configuration. Then we investigated the maximum reach for different spectral efficiencies, after nonlinear propagation over uncompensated links with lumped amplification. Performance for systems based on both standard single-mode fiber (SSMF) and large effective area non-zero dispersion-shifted fiber (NZDSF) has been analyzed. Assuming SSMF with 25-dB span loss, we found that PM-BPSK can reach 6480 km at a net capacity of 4 Tb/s across the C band. Conversely, PM-16QAM can deliver 27 Tb/s, but over 270 km only. Note that a lower span length, the use of Raman amplification and/or pure silica-core fibers (PSCFs) can significantly increase the maximum reach, but without changing the hierarchy among the performance of modulation formats. We also show that the maximum reachable distance is approximately 2/3 of the one achievable in linear propagation at the optimum launch power, regardless of the modulation format, spacing and fiber type. As additional results, we also verified that the optimum launch power per subcarrier linearly depends on the span loss, varies with the fiber type, but it is independent of the modulation format, and that the relationship between the maximum reachable distance and the span loss is almost linear.

On the Performance of Nyquist-WDM Terabit Superchannels Based on PM-BPSK, PM-QPSK, PM-8QAM or PM-16QAM Subcarriers

We address perturbative models for the impact of nonlinear propagation in uncompensated links. We concentrate on a recently-proposed model which splits up the signal into spectral components and then resorts to a four-wave-mixing-like approach to assess the generation of nonlinear interference due to the beating of the signal spectral components. We put its founding assumptions on firmer ground and we provide a detailed derivation for its main analytical results. We then carry out an extensive simulative validation by addressing an ample and significant set of formats encompassing PM-BPSK, PM-QPSK, PM-8QAM, and PM-16QAM, all operating at 32 GBaud. We compare the model prediction of maximum system reach and optimum launch power versus simulation results, for all four formats, three different kinds of fibers (PSCF, SMF, and NZDSF) and for several values of WDM channel spacing, ranging from 50 GHz down to the symbol-rate. We found that, throughout all tests, the model delivers accurate predictions, potentially making it an effective general-purpose system design tool for coherent uncompensated transmission systems.

Modeling of the Impact of Nonlinear Propagation Effects in Uncompensated Optical Coherent Transmission Links

Digital signal processing has completely changed the way optical communication systems work during recent years. In combination with coherent demodulation, it enables compensation of optical distortions that seemed impossible only a few years ago. However, at high bit rates, this comes at the price of complex processing circuits and high power consumption. In order to translate theoretic concepts into economically viable products, careful design of the digital signal processing algorithms is needed. In this paper, we give an overview of digital equalization algorithms for coherent receivers and derive expressions for their complexity. We compare single-carrier and multicarrier approaches, and investigate blind equalizer adaptation as well as training-symbol-based algorithms. We examine tradeoffs between parameters like sampling rate and tracking speed that are important for algorithm design and practical implementation.

Equalizer Design and Complexity for Digital Coherent Receivers

We generate 56-Gbaud PDM 16-QAM using electronic time-division multiplexing (ETDM) and a four-level-driven I/Q modulator. The 448-Gb/s line-rate signal is transmitted over 1,200 km of ultra-large-area fiber and coherently received by two 32.5-GHz oscilloscopes with >5.5 effective bits.

Generation and 1,200-km transmission of 448-Gb/s ETDM 56-Gbaud PDM 16-QAM using a single I/Q modulator

Modulation formats for 100G and beyond

The rise of coherent detection and digital signal processing is drastically changing the design of optical transmission systems. In this paper we review the challenges and opportunities offered by such receivers in the design of long-haul 100G systems.

/pdf/polmux-qpsk-modulation-and-coherent-detection-the-challenge-49wf7yx4ko.pdf

POLMUX-QPSK modulation and coherent detection: The challenge of long-haul 100G transmission

We demonstrate the generation and transmission of 11 channels with 224-Gb/s polarization-multiplexed return-to-zero 16-level quadrature amplitude modulation over 670 km of standard single-mode fiber (SSMF) with 50-GHz channel spacing and a spectral efficiency of 4.2 b/s/Hz. We report a penalty of around 4.3 dB in the performance at back-to-back in comparison to the theoretical limits, and a margin of 1 dB in Q-factor below the forward-error correction limit (assumed to be at a bit-error rate of 3.8 × 10-3) after transmission over 670 km of SSMF.

/pdf/11-times-224-gb-s-polmux-rz-16qam-transmission-over-670-km-1zjz1hr8i1.pdf

11 $\,\times\,$ 224-Gb/s POLMUX-RZ-16QAM Transmission Over 670 km of SSMF With 50-GHz Channel Spacing

We investigate, by means of simulations, polarization-multiplexed (PolMux) quadrature duobinary (QDB) as a modulation format for transmitting 111-Gb/s signals with high spectral efficiency (SE) and acceptable system complexity. We show that PolMux-QDB can be used to transmit 111-Gb/s signals with an SE of 4-b/s/Hz, but at the same time has a higher tolerance to nonlinear transmission effects and local oscillator laser linewidth compared to PolMux-16QAM at the same data rate and SE.

/pdf/111-gb-s-polmux-quadrature-duobinary-for-robust-and-4319r2d3i1.pdf

111-Gb/s PolMux-Quadrature Duobinary for Robust and Bandwidth Efficient Transmission

We demonstrate transmission of 11 × 224-Gb/s POLMUX-RZ-16QAM over 1500 km with a channel spacing of 50 GHz. A hybrid configuration of LongLine and pure silica fiber is used to optimize both nonlinear tolerance and Raman gain.

/pdf/transmission-of-11-224-gb-s-polmux-rz-16qam-over-1500-km-of-2btmcnyy09.pdf

Transmission of 11 × 224-Gb/s POLMUX-RZ-16QAM over 1500 km of LongLine and pure-silica SMF

The authors demonstrate the successful transmission of 10 channels with 224-Gb/s POLMUX-16QAM modulation (28 GBaud) on a 37.5-GHz wavelength grid. Using large-Aeff pure-silica-core fibers they show a 656-km transmission distance with a spectral efficiency of 5.6 b/s/Hz. They report a back-to-back performance penalty of 3.5 dB compared to theoretical limits at the forward-error correction (FEC) limit (bit-error rate of 3.8·10-3), and a margin of 0.5 dB in Q-factor with respect to the FEC limit after 656 km of transmission.

/pdf/10-times-224-gb-s-polmux-16qam-transmission-over-656-km-of-47n7epksp1.pdf

T. Wuth

Papers

POLMUX-QPSK modulation and coherent detection: The challenge of long-haul 100G transmission

11 $\,\times\,$ 224-Gb/s POLMUX-RZ-16QAM Transmission Over 670 km of SSMF With 50-GHz Channel Spacing

111-Gb/s PolMux-Quadrature Duobinary for Robust and Bandwidth Efficient Transmission

Transmission of 11 × 224-Gb/s POLMUX-RZ-16QAM over 1500 km of LongLine and pure-silica SMF

10 $\times$ 224-Gb/s POLMUX-16QAM Transmission Over 656 km of Large- ${\rm A}_{\rm eff}$ PSCF With a Spectral Efficiency of 5.6 b/s/Hz