Replacing the Soft FEC Limit Paradigm in the Design of Optical Communication Systems

Abstract: The FEC limit paradigm is the prevalent practice for designing optical communication systems to attain a certain bit-error rate (BER) without forward error correction (FEC). This practice assumes that there is an FEC code that will reduce the BER after decoding to the desired level. In this paper, we challenge this practice and show that the concept of a channel-independent FEC limit is invalid for soft-decision bit-wise decoding. It is shown that for low code rates and high order modulation formats, the use of the soft FEC limit paradigm can underestimate the spectral efficiencies by up to 20%. A better predictor for the BER after decoding is the generalized mutual information, which is shown to give consistent post-FEC BER predictions across different channel conditions and modulation formats. Extensive optical full-field simulations and experiments are carried out in both the linear and nonlinear transmission regimes to confirm the theoretical analysis.

• The paper demonstrates that traditional soft FEC limits can underestimate spectral efficiencies by up to 20%, advocating GMI as a more reliable predictor.
• The paper details simulations and experimental validations that reveal pre-FEC BER and MI shortcomings in accurately predicting post-FEC performance.
• The paper highlights that adopting GMI streamlines optical system design by ensuring consistent performance metrics across various modulation formats.

Evaluating the Soft FEC Limit Paradigm in Optical Communication Systems Design

The paper "Replacing the Soft FEC Limit Paradigm in the Design of Optical Communication Systems" by Alvarado et al. critically examines a prevalent practice in optical communications—designing systems to a "soft FEC limit" paradigm. This paradigm assumes that if a system achieves a certain pre-forward error correction (FEC) bit-error rate (BER), the implemented FEC can reliably reduce this BER to an acceptable post-FEC level.

Summary of the Problem and Approach

Modern optical communication systems often employ soft-decision FEC (SD-FEC) in combination with multilevel modulation formats, collectively referred to as coded modulation. While the SD-FEC paradigm has gained popularity due to its potential for higher spectral efficiencies, the paper critiques its reliance on the pre-FEC BER as a predictor for post-FEC performance, particularly across different modulation formats and code rates.

The authors challenge the validity of this approach, demonstrating that the soft FEC limit paradigm can underestimate spectral efficiencies by up to 20%, especially for lower code rates and higher-order modulation formats. Instead, the study promotes the generalized mutual information (GMI) as a more robust and consistent predictor for post-FEC BER.

Key Findings and Numerical Results

The study meticulously evaluates the predictive accuracy of pre-FEC BER, mutual information (MI), and GMI:

1. Pre-FEC BER and Limitations: The pre-FEC BER, while historically used as a straightforward benchmark, does not reliably predict post-FEC BER for SD-FEC. The assumption that similar pre-FEC BERs will yield similar post-FEC performance across different constellations is shown to be flawed, particularly at low and medium code rates.
2. Mutual Information: MI provides certain insights but is shown not to be comprehensive enough when others, like unconventional constellations (e.g., 8QAM), are considered. The study underscores that MI's indifference to bit-labeling adversely impacts its predictive capacity for SD-FEC.
3. Generalized Mutual Information: GMI emerges as a highly accurate post-FEC BER predictor, demonstrating uniformity across various modulation formats and maintaining consistency in both linear and nonlinear transmission regimes. Simulation and experimental validation underline GMI's utility, showing robustness even at high spectral efficiencies typical in modern optical systems.

Theoretical and Practical Implications

The paper's implication extends to both theoretical and practical realms of optical communications. Theoretically, it integrates the GMI as a foundational metric, encouraging a shift from traditional practices centered on the soft FEC limit. Practically, the adoption of GMI could streamline the design processes of optical systems, ensuring a more consistent measure of system performance across diverse configurations.

Since the results reveal substantial underestimations when using traditional paradigms, adopting a GMI-based design criterion could lead to enhancements in system capacity and efficiency, bringing tangible benefits in areas relying on optical fiber communications.

Future Prospects in Optical Communication Systems

This analysis encourages future research to explore the universality of GMI across different system implementations, including more advanced FEC schemes and non-uniform signal constellations. Moreover, the exploration of GMI in conjunction with other advanced decoding techniques, such as nonbinary or iterative decoders, could further refine post-FEC BER predictions. Additionally, the extension of such analyses to hybrid optical-electrical channels could offer insights into optimizing integrated communication network designs.

In conclusion, the research presented by Alvarado et al. brings attention to the limitations of traditional FEC paradigms in optical communications and introduces GMI as a viable alternative for better system design and performance prediction. By aligning closer with theoretical capacities, GMI holds the potential to redefine efficiency benchmarks in optical communication.