🔬 optics

Reducing Turbulence-Induced Outages in a Deployed Terrestrial Free-Space Optical Communication Link via Interleaving

This paper presents an experimental study on a 4.6 km urban terrestrial free-space optical link demonstrating that data interleaving reduces turbulence-induced outage probability by two orders of magnitude while revealing a critical dependency between turbulence strength, interleaver length, and achievable data rate for robust system design.

Original authors: Kadir Gümüş, Vincent van Vliet, Menno van den Hout, Thomas Bradley, Eduward Tangdiongga, Chigo Okonkwo

Published 2026-06-16

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Kadir Gümüş, Vincent van Vliet, Menno van den Hout, Thomas Bradley, Eduward Tangdiongga, Chigo Okonkwo

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

Imagine you are trying to shout a long, complex story to a friend across a busy, windy city park. This is essentially what Free-Space Optical (FSO) communication does: it sends data using invisible beams of light through the air instead of fiber-optic cables.

However, just like your voice, the light beam gets messed up by the atmosphere. The paper describes a 4.6-kilometer "shout" between two university campuses in Eindhoven, Netherlands.

Here is the breakdown of their experiment and findings using simple analogies:

The Problem: The "Windy Day" Effect

The main villain in this story is atmospheric turbulence. Think of the air not as empty space, but as a soup of hot and cold pockets (like the air rising off a hot asphalt road). As the light beam travels through these pockets, it gets bent and scattered.

The Result: Sometimes the light hits your friend's eye clearly; other times, the "wind" blocks it completely for a split second.
The Consequence: In data terms, this causes "outages" (the connection drops) or "burst errors" (a whole chunk of the story is garbled at once). If you are trying to send high-speed video or data, these sudden drops are disastrous.

The Solution: The "Shuffle" Trick (Interleaving)

The researchers tested a technique called data interleaving. To understand this, imagine you are writing a letter to your friend, but you know the wind might blow away a specific page.

Without Interleaving: You write the letter normally: "Dear Friend, I hope you are well..." If the wind blows away the middle of the page, your friend gets: "Dear Friend, [BLANK] hope you are well..." The sentence is broken and hard to fix.
With Interleaving: You write the letter, but you shuffle the words before sending them. You send the first word of the first sentence, then the first word of the second sentence, then the first word of the third, and so on.
- If the wind blows away a chunk of the transmission, it doesn't take a whole sentence. Instead, it takes one word from Sentence A, one from Sentence B, and one from Sentence C.
- When your friend receives the message, they un-shuffle it. Because the missing words are spread out, the computer (or your friend's brain) can easily guess what the missing words were based on the surrounding context.

What They Did

The team set up a real-world testbed between two campuses. They measured how "windy" (turbulent) the air was over a month and recorded how much the light signal fluctuated. They then simulated sending data through this exact environment using different "shuffle" speeds (interleaver lengths).

The Key Findings

The Magic of Shuffling: Using this "shuffle" technique reduced the chance of a total connection failure (outage) by 100 times (two orders of magnitude). That is a massive improvement.
The Trade-Off: The "windier" the day (stronger turbulence), the more you need to shuffle.
- Calm Day: You don't need to shuffle much.
- Stormy Day: You need a very long shuffle (a larger buffer) to spread the errors out enough to fix them.
The Cost: The downside of shuffling is latency (delay). Just like it takes time to shuffle a deck of cards and then un-shuffle them, the computer has to wait until it has collected enough data to start the shuffle. The stronger the turbulence, the longer you have to wait, which slows down the immediate response time.

The Bottom Line

The paper proves that for high-speed light communication in cities, you cannot just rely on the raw signal. You must use a "shuffling" strategy to survive the atmospheric turbulence.

If the air is calm: You can send data very fast with almost no delay.
If the air is turbulent: You must accept a slightly longer delay (by using a longer "shuffle") to keep the data rate high and the connection reliable.

This study provides a rulebook for engineers: Measure the "wind" (turbulence), and then choose the right "shuffle" size to keep the connection alive.

Technical Summary: Reducing Turbulence-Induced Outages in a Deployed Terrestrial Free-Space Optical Communication Link via Interleaving

Problem Statement
Terrestrial Free-Space Optical (FSO) communication offers high-data-rate wireless connectivity but faces significant reliability challenges due to atmospheric turbulence. Unlike satellite links, terrestrial beams traverse the dense lower atmosphere, causing severe fluctuations in received optical power and irradiance. These fluctuations degrade the Signal-to-Noise Ratio (SNR) over time, leading to high outage probabilities that hinder widespread deployment. While data interleaving has been proposed to mitigate burst errors caused by turbulence by redistributing them across multiple codewords, existing experimental works on terrestrial links have not sufficiently analyzed how outage probability is affected by interleavers under realistic urban atmospheric conditions.

Methodology
The authors conducted an experimental study using the Reid Photonloop testbed, a 4.6 km urban FSO link connecting the Eindhoven University of Technology Campus to the High Tech Campus. The experimental setup involved:

Power Measurement: An external cavity laser (1542 nm) amplified to 29 dBm was transmitted via fiber-coupled terminals. Received power ( $P_{rec}$ ) was measured at 10 µs intervals using a high-speed power meter over a period from October 10th to 13th, 2025.
Turbulence Characterization: A co-located Scintec BLS900 Neo scintillometer measured the refractive index structure parameter ( $C_n^2$ ) with a 10-minute averaging time to characterize optical turbulence.
Data Analysis: The study focused on three specific turbulence regimes identified from the $C_n^2$ distribution: weak ( $2.19 \cdot 10^{-16} \text{ m}^{-2/3}$ ), moderate ( $1.62 \cdot 10^{-15} \text{ m}^{-2/3}$ ), and strong ( $6.77 \cdot 10^{-15} \text{ m}^{-2/3}$ ).
Simulation Model: The authors simulated a system using 256-QAM at a 50 GBaud symbol rate, employing punctured Low-Density Parity-Check (LDPC) codes from the DVB-S2 standard (blocklength $N=64800$ bits). They modeled a symbol-wise block interleaver/deinterleaver to evaluate the impact of different interleaver lengths (20 ms, 50 ms, 100 ms) on outage probability ( $P_{out}$ ).

Key Contributions and Results
The paper presents the following findings based on the experimental data and simulations:

Turbulence Correlation: A direct correlation was observed between measured $C_n^2$ and the severity of $P_{rec}$ fluctuations. Higher turbulence levels resulted in more severe power fades and SNR variations.
Interleaving Efficacy: The study demonstrates that interleaving can reduce outage probability by two orders of magnitude.
- In weak turbulence, throughput remained consistent over time, and interleaving had a negligible effect on $P_{out}$ .
- In moderate turbulence, interleaving enabled up to a 70% higher decoded data rate compared to no interleaving when $P_{out} < 10^{-2}$ . A 20 ms interleaver significantly reduced outage probability, with longer lengths providing incremental gains.
- In strong turbulence, an interleaver of at least 50 ms was required to achieve $P_{out} < 10^{-2}$ at decoded data rates above 200 Gb/s.
Design Trade-offs: The results reveal a clear dependency among turbulence strength, interleaver length, and achievable data rate. While longer interleavers improve reliability in high-turbulence conditions, they introduce latency equal to $2t$ (where $t$ is the interleaver length), as the matrix must be filled before transmission or decoding can proceed.

Significance
The authors claim that this work fills a critical gap by analyzing interleaving in a deployed terrestrial FSO system under realistic urban conditions. The study establishes a direct link between atmospheric conditions ( $C_n^2$ ) and interleaver design requirements. Specifically, it provides practical design insights indicating that as turbulence strength increases, longer interleavers are necessary to sustain high data rates at low outage probabilities. This relationship forms a foundation for developing adaptive interleaving strategies in future terrestrial FSO systems, moving beyond theoretical models to experimentally validated design parameters.