Mitigation of nonlinearities using conjugate data repetition

Henrik Eliasson, Pontus Johannisson, Magnus Karlsson, Peter A. AndreksonOptics Express 2015 · DOI

TL;DR

This paper introduces "conjugate data repetition" (CDR), a novel technique to combat signal degradation caused by fiber optic nonlinearities. By transmitting a signal alongside a time-shifted, conjugated copy, CDR effectively doubles the transmission reach of optical communication systems compared to standard methods, offering a promising path towards higher-capacity fiber networks.

What Problem Does This Paper Address?

Modern high-speed fiber optic communication systems face a critical bottleneck: signal distortion caused by nonlinear effects within the optical fiber. These nonlinearities, primarily driven by the Kerr effect, become more pronounced with increasing optical power, limiting the signal-to-noise ratio and thus the maximum transmission distance. While digital back-propagation offers a solution, it demands significant computational resources. Other techniques like phase-conjugated twin waves (PCTW) have shown promise but often come at the cost of reduced spectral efficiency. This paper addresses the need for a method that effectively mitigates nonlinearities without drastically sacrificing spectral efficiency or requiring excessive computational power.

Methods

The core of this paper lies in the introduction and theoretical analysis of conjugate data repetition (CDR). Unlike PCTW, where a conjugated copy is transmitted on an orthogonal polarization, CDR transmits the conjugated copy in a different time slot. This temporal separation offers several advantages:

  • Simplified Digital Signal Processing (DSP): CDR is compatible with the conventional Constant Modulus Algorithm (CMA) for adaptive channel equalization, unlike PCTW which requires a more complex polarization-switched CMA. This makes integration into existing PM-QPSK hardware potentially easier.
  • Theoretical Analysis: The authors provide a theoretical explanation for CDR's nonlinearity mitigation capabilities using a time-domain perturbation analysis. This analysis focuses on how specific "pulse triplets" (combinations of signal and conjugated pulses) generate nonlinear distortions that cancel out when the received signal and its conjugated copy are linearly superposed.
  • Numerical Simulations: The theoretical findings are validated through extensive numerical simulations. These simulations compare the transmission reach of CDR-QPSK and CDR-16QAM against PM-QPSK and PCTW-QPSK under various conditions, including single-channel and WDM systems, and with and without inline ASE noise.

Datasets

Not discussed in the paper. The evaluation is based on numerical simulations of optical fiber transmission systems.

Key Insights and Ideas

  • Temporal Separation is Key: The novel aspect of CDR is transmitting the conjugated copy in a different time slot. This seemingly simple change unlocks significant advantages in DSP complexity and implementation.
  • CMA Compatibility: The ability to use the conventional CMA for equalization is a major practical advantage of CDR over PCTW, potentially simplifying hardware upgrades.
  • Theoretical Cancellation Mechanism: The perturbation analysis reveals that specific combinations of signal and conjugated pulses, when transmitted with an antisymmetric dispersion map and symmetric power map, generate nonlinear distortions that cancel each other out upon coherent superposition in the receiver.
  • Surprising Performance: Despite the perturbation analysis not fully explaining how PCTW outperforms CDR, simulations show remarkably similar performance, suggesting CDR is a highly effective nonlinearity mitigation technique.
  • Dispersion Pre-compensation Nuance: The simulations reveal that the optimal dispersion pre-compensation for CDR is not precisely 50% as predicted by first-order theory, but closer to 40%, likely due to signal-noise interactions. However, the penalty for using 50% is negligible, allowing for simpler symmetric dispersion maps.

Key Results

  • Doubled Transmission Reach: In single-channel systems, CDR-QPSK achieves approximately twice the transmission reach of PM-QPSK at the same channel bit rate.
  • WDM Performance: In a 7-channel WDM system, CDR-QPSK offers a reach increase of approximately 1.8 times compared to PM-QPSK.
  • PCTW Comparison: CDR-QPSK demonstrates performance very similar to PCTW-QPSK, with a slightly higher reach increase of approximately 2.1 times in single-channel systems.
  • Spectral Efficiency Trade-off: While CDR-QPSK offers improved reach, it halves the spectral efficiency compared to PM-QPSK. Attempts to use CDR with 16QAM to maintain spectral efficiency did not yield significant improvements in transmission reach.
  • Nonlinear Phase Shift Tolerance: CDR-QPSK exhibits a significantly increased tolerance to nonlinear phase shifts, being more than five times greater than PM-QPSK.

Limitations and Future Work

  • Spectral Efficiency: The primary limitation of CDR-QPSK is its reduced spectral efficiency (1 bit/s/Hz) compared to PM-QPSK (2 bit/s/Hz). While CDR-16QAM was explored to maintain SE, it did not yield significant reach improvements.
  • Dispersion Map Optimization: While the theory suggests a symmetric dispersion map, simulations indicate that optimal performance might be achieved with approximately 40% dispersion pre-compensation, with the deviation attributed to signal-noise interactions. However, the penalty for using a symmetric map is minimal.
  • PMD Neglected: Polarization Mode Dispersion (PMD) was neglected in the simulations, which could impact performance in real-world systems. The authors note that previous work suggests PCTW can improve PMD tolerance.
  • Future Work: The authors suggest further investigation into the impact of PMD on CDR systems and exploring alternative modulation formats or coding schemes to improve spectral efficiency. The potential for Raman amplification to achieve a more symmetric power map is also mentioned as a future avenue.

Summary updated

Abstract

We investigate a time-domain implementation of generalized phase-conjugated twin waves which we call conjugate data repetition. A theory based on time-domain perturbation analysis explaining the mitigation of nonlinear effects is provided, and the concept is evaluated using numerical simulations. Compared to PM-QPSK at the same channel bit rate, the single-channel transmission reach in a conventional system with standard single-mode fiber of conjugate data repetition-QPSK is increased by approximately a factor of 2.
BibTeX
@inproceedings{eliasson2015mitigation,
  author = {Henrik Eliasson and Pontus Johannisson and Magnus Karlsson and Peter A. Andrekson},
  title = {{Mitigation of nonlinearities using conjugate data repetition}},
  booktitle = {Optics Express},
  year = {2015},
  doi = {10.1364/oe.23.002392},
}
Figure