🔬 optics

Gas Line Absorption Mitigation in Hollow-Core Fibre using Spectral Pre-Equalisation

This paper demonstrates that using spectral pre-equalization to mitigate $\text{CO}_2$ absorption in hollow-core fibers can reduce the Q-factor penalty by 5.5 dB, outperforming traditional digital post-equalization methods.

Original authors: Eric Sillekens, Ronit Sohanpal

Published 2026-02-11

📖 3 min read☕ Coffee break read

CC BY 4.0

Original authors: Eric Sillekens, Ronit Sohanpal

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

The Problem: The "Ghost in the Glass"

Imagine you are trying to listen to a high-fidelity concert through a long, clear glass tube. Usually, this would be perfect. But in this specific type of tube—called a Hollow-Core Fibre (HCF)—the inside isn't a vacuum; it’s filled with air.

The problem is that certain gases, specifically Carbon Dioxide (CO2), act like tiny, invisible "sound sponges." As the music (your data) travels through the tube, these CO2 molecules "soak up" specific notes. If you are playing a beautiful symphony, the CO2 might suddenly swallow all the middle-C notes. The music becomes distorted, choppy, and hard to understand.

In the world of high-speed internet, this "swallowing of notes" is called Gas-Line Absorption (GLA). It creates "holes" in the signal, making it incredibly difficult for the receiver at the other end to make sense of the data.

The Old Way: The "Brute Force" Fix

To fix this, engineers usually use a digital tool called an Equaliser. Think of this like a very smart person at the end of the tube trying to "guess" what the missing notes were by listening to the surrounding sounds.

However, because the CO2 "holes" are so sharp and sudden, this person has to work incredibly hard. They need a massive, complex mental checklist (in technical terms, 383 "taps") to try and reconstruct the missing pieces. It’s like trying to solve a massive, complicated jigsaw puzzle in real-time while the pieces are still flying at you. It works, but it requires a massive amount of "brainpower" (computational energy), which is expensive and slow.

The New Way: The "Pre-Emphasis" Trick

The researchers in this paper proposed a much cleverer solution: Spectral Pre-Equalisation.

Instead of trying to fix the holes after the music has already been ruined, they decide to "over-emphasize" the notes before they even enter the tube.

The Analogy: Imagine you know you are about to walk through a room filled with people who love to steal hats. Instead of walking in normally and then trying to buy a new hat later, you decide to wear a giant, heavy, reinforced helmet. You "over-compensate" for the loss before it even happens.

In technical terms, they look at the "holes" the gas makes, and they digitally boost the signal in those specific frequencies before it travels through the fibre. By the time the signal hits the CO2 "sponges," the signal is so "loud" in those specific spots that even after the gas soaks some up, the remaining signal is still clear and strong.

Why This Matters (The Results)

The researchers found that this "pre-emphasis" trick is a massive win for two reasons:

Better Quality: It actually works better than the brute-force method. It recovered 5.5 dB of signal quality, beating the old method by 1.3 dB. In the world of high-speed data, that’s a significant boost in reliability.
Extreme Efficiency: This is the real kicker. While the old method needed a massive "brain" with 383 taps to fix the error, the new method only needs a tiny "brain" with 3 taps.

The Bottom Line: It’s like replacing a massive, power-hungry supercomputer with a simple, elegant mathematical shortcut. This makes it much easier and cheaper to build the next generation of ultra-fast, long-distance internet cables that use hollow-core technology.

Technical Summary: Gas Line Absorption Mitigation in Hollow-Core Fibre using Spectral Pre-Equalisation

Problem: Gas-Line Absorption (GLA) in HCF

Hollow-core fibres (HCFs) offer transformative advantages for optical communications, including ultra-low attenuation, reduced latency, and low chromatic dispersion. However, a critical physical limitation is Gas-Line Absorption (GLA). This phenomenon is caused by residual atmospheric gases (primarily $\text{CO}_2$ , but also $\text{CO}$ and $\text{H}_2\text{O}$ ) trapped within the hollow core.

$\text{CO}_2$ induces sharp, localized absorption peaks (notches) across the C-, L-, and U-bands. These notches cause significant spectral impairments, specifically increasing inter-symbol interference (ISI) and degrading the Signal-to-Noise Ratio (SNR). In high-baud-rate systems, this can lead to a Q-factor penalty of up to 6 dB, severely limiting the usable bandwidth and transmission distance of HCF-based links.

Methodology

The researchers employed a simulation-based approach to model and mitigate these impairments:

Modeling: They simulated a 300 km HCF link using a 140 GBd dual-polarization 1024-QAM signal. The GLA was modeled as a series of Lorentzian-shaped notches in the frequency domain. To ensure physical realism, they enforced causality by using the Hilbert transform to reconstruct the phase response from the amplitude notches.
Baseline Mitigation (Adaptive Equalization): The study first evaluated standard digital signal processing (DSP) using a Recursive Least Squares (RLS) adaptive equalizer to see how many "taps" (computational steps) would be required to compensate for the ISI caused by the notches.
Proposed Mitigation (Spectral Pre-Equalisation): The authors proposed a novel frequency-domain pre-equalization scheme. The process involves:
- Estimating the Power Spectral Density (PSD) of the received signal using Welch’s method.
- Calculating a target frequency response by taking the inverse square-root of the PSD to flatten the spectrum.
- Reconstructing the phase using a negated Hilbert transform to ensure a unity net phase response.
- Applying a regularizer ( $\lambda$ ) to the pre-equalizer to balance the trade-off between ISI mitigation and noise enhancement (preventing the system from over-amplifying noise in the absorption notches).

Key Contributions

Characterization of GLA Impact: Quantified the relationship between notch depth, equalizer tap length, and Q-factor degradation in high-baud-rate coherent systems.
Novel DSP Scheme: Introduced a spectral pre-equalization technique that targets the frequency-domain characteristics of gas absorption before the signal reaches the adaptive equalizer.
Complexity Reduction: Demonstrated that frequency-domain pre-equalization significantly reduces the computational burden on the subsequent time-domain adaptive equalizer.

Results

The simulation results demonstrated the superiority of the proposed method over traditional adaptive equalization:

Performance Gain: For a 10 dB notch depth, the spectral pre-equalization scheme achieved a 5.5 dB reduction in Q-factor penalty.
Efficiency vs. Complexity:
- To recover 4.2 dB of Q-factor degradation using only an RLS equalizer, a massive 383 taps were required.
- The proposed pre-equalization scheme outperformed the 383-tap RLS equalizer by 1.3 dB while requiring only 3 adaptive taps.
Limitations: The scheme cannot achieve perfect mitigation because the absorption notches inherently reduce the OSNR within those specific frequency bands, and the estimation process introduces a small amount of noise.

Significance

This work is highly significant for the future deployment of long-haul HCF networks. While physical methods like "purging" the gas from the fiber exist, they are slow and impractical for long distances. This research proves that DSP-based mitigation is a highly effective, flexible, and computationally efficient alternative. By reducing the complexity required to handle GLA, this method paves the way for high-capacity, high-speed optical transmission through hollow-core fibers, maximizing their potential for next-generation telecommunications.