⚡ electrical engineering

Statistical Characterization of Wind-Induced Beam Refraction in Water-to-Air Optical Channels

This paper proposes a novel statistical channel model for vertical water-to-air optical links that accounts for wind-induced beam refraction and pointing errors using a Beta mixture distribution, deriving closed-form expressions for channel distribution and outage probability while revealing a fundamental reliability floor caused by total internal reflection.

Original authors: Mohamed Nennouche, Ikenna Chinazaekpere Ijeh, Mohammad-Ali Khalighi

Published 2026-05-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Mohamed Nennouche, Ikenna Chinazaekpere Ijeh, Mohammad-Ali Khalighi

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 Big Picture: Sending a Flashlight from the Deep to the Sky

Imagine you are underwater, deep in the ocean, holding a powerful laser pointer. Your goal is to shine that light straight up to a drone (a UAV) hovering in the sky to send it a secret message.

This sounds simple, but the ocean surface is not a calm, flat mirror. It is a wavy, moving dance floor. Every time a wave rolls by, it tilts the "mirror" of the water. This tilting bends the light beam (a phenomenon called refraction) just like a straw looks bent in a glass of water.

The paper is a guidebook for engineers trying to figure out: "How likely is it that our light beam will actually hit the drone's camera, given that the ocean is constantly shaking?"

The Three Main Problems

The authors identified three main ways this communication link can fail:

The "Wobbly Mirror" Effect (Pointing Errors):
Because the waves are moving, the laser beam doesn't always go straight up. It gets knocked off course.
- The Analogy: Imagine trying to throw a ball through a hoop held by a friend who is standing on a trampoline. Even if you aim perfectly, the trampoline's bounce might send the ball flying to the left or right, missing the hoop entirely.
- The Paper's Solution: The authors created a mathematical "weather forecast" for this wobbling. They used a complex statistical tool (called a Beta Mixture Model) to predict exactly how much the beam will wobble based on how windy it is. They found that wind speed is the main driver of this wobble.
The "Total Internal Reflection" Trap (The Glass Ceiling):
Sometimes, the waves get so steep that the light doesn't escape the water at all. Instead of going up, the light hits the bottom of the wave at such a sharp angle that it bounces back down into the ocean.
- The Analogy: Think of a swimmer looking up at the surface of a pool. If they look at a very shallow angle, the water surface looks like a shiny silver mirror, and they can't see the sky. The light gets trapped underwater.
- The Paper's Finding: This creates a "hard floor" for reliability. No matter how good your equipment is, if the wind is strong enough to create steep waves, a certain percentage of your messages will never get out because the physics of light traps them.
The "Flashlight Beam" Problem (Field of View):
Even if the light escapes the water, it might miss the drone's camera if the camera isn't looking in the right direction or doesn't have a wide enough "eye" (Field of View).
- The Analogy: If you are trying to catch a fly with a net, but the fly is buzzing around wildly and your net is very small, you will miss it. The drone needs a wide enough "net" (camera lens) to catch the wandering light beam.

The Surprising Discovery: Going Deeper Helps?

One of the most interesting findings in the paper is counter-intuitive. Usually, in underwater communication, going deeper is bad because the water absorbs the light, making the signal weaker.

However, the authors found that going deeper actually makes the connection more stable in this specific scenario.

The Analogy: Imagine shining a flashlight through a foggy room. If you are close to the wall, a tiny shake of your hand makes the dot on the wall jump wildly. But if you stand far back, the beam spreads out into a wide, soft glow. A tiny shake of your hand barely moves the center of that big, soft glow.
The Result: By placing the underwater laser deeper (e.g., 30 meters instead of 10 meters), the beam spreads out more by the time it hits the surface. This "wide glow" is harder for the waves to knock off-target. The slight loss of brightness due to depth is worth the gain in stability.

The Bottom Line

The paper provides a new, easy-to-use formula for engineers to design these underwater-to-air links.

It predicts failure: It tells you exactly how often the link will break based on wind speed and wave height.
It sets a limit: It proves that there is a physical limit to how reliable this can be because of the "trapped light" effect (Total Internal Reflection). You can't engineer your way out of physics; if the waves are too steep, the light stays underwater.
It offers a design tip: To make the link more reliable, you might actually want to put your underwater transmitter deeper to let the beam spread out, rather than keeping it shallow.

In short, the paper turns a chaotic, wavy ocean problem into a predictable math problem, helping engineers build better systems to talk between the deep sea and the sky.

Technical Summary: Statistical Characterization of Wind-Induced Beam Refraction in Water-to-Air Optical Channels

Problem Statement
Direct water-to-air (W2A) optical communication is a critical enabler for the Internet of Underwater Things (IoUT), offering high-speed data retrieval without the need for intermediate surface platforms. However, the dynamic nature of the sea surface introduces significant channel complexity. Wave-induced random refraction and reflection cause severe pointing errors (PEs) and potential link interruptions. While previous studies have relied on computationally expensive Monte-Carlo (MC) simulations or addressed air-to-water links, there is a lack of tractable, closed-form statistical models for vertical W2A links that account for wind speed, total internal reflection (TIR), and receiver field-of-view (FoV) limitations.

Methodology
The authors propose a comprehensive statistical channel model for a vertical link between an underwater base station and an unmanned aerial vehicle (UAV). The system assumes a static transmitter and a hovering UAV, identifying the random motion of the sea surface as the dominant source of PEs.

Geometric and Physical Modeling: The model defines the incident angle ( $\theta_I$ ) at the sea surface and the resulting angle-of-arrival ( $\theta_A$ ) in the air. The incident angle is modeled using a Weibull distribution, with shape and scale parameters derived as functions of wind speed ( $U$ ).
Statistical Approximation of Pointing Errors: Due to the non-linear geometric transformation between $\theta_I$ and $\theta_A$ , deriving a closed-form probability density function (PDF) for the radial displacement is analytically intractable. The authors evaluate several candidate distributions and find that $\theta_A$ is best approximated by a Weibull distribution.
Beta Mixture Model (BMM): To characterize the resulting pointing error loss ( $h_P$ ), the authors employ a novel statistical approach using a finite Beta mixture model. The parameters of this mixture are optimized using the Expectation–Maximization (EM) algorithm. This approach provides a tractable, closed-form expression for the channel distribution, overcoming the complexity of direct analytical derivation.
Link Interruption Analysis: The model explicitly accounts for link interruptions caused by two mechanisms:
- Total Internal Reflection (TIR): Occurs when the incident angle exceeds the critical angle ( $\theta_c \approx 48.75^\circ$ ).
- Geometric Misalignment: Occurs when the refracted beam falls outside the receiver's FoV.
  The probability of these interruptions is derived analytically.
Outage Probability Derivation: Combining the path loss, the BMM-based pointing error loss, and the interruption coefficient, the authors derive a closed-form expression for the overall channel PDF and the link outage probability ( $P_{out}$ ).

Key Contributions

Novel Tractable Model: The paper introduces a closed-form statistical channel model for W2A links that incorporates wind speed effects via a Beta mixture distribution fitted to pointing errors.
Analytical Outage Expressions: The authors derive exact closed-form expressions for the channel distribution and link outage probability, explicitly accounting for TIR and FoV constraints.
Identification of Outage Floor: The analysis reveals a fundamental "outage floor" imposed by TIR, which limits the minimum achievable outage probability regardless of other system optimizations.
Performance Insights: The study provides analytical insights into how system parameters (UAV altitude, transmitter depth, and FoV) interact with wind speed to affect link reliability.

Results
The analytical results were validated against extensive MC simulations, showing high accuracy (e.g., $R^2 > 0.97$ for the BMM fit). Key findings include:

Wind Speed Impact: Higher wind speeds significantly increase the probability of pointing errors and link interruptions.
Altitude vs. Depth: Increasing the UAV altitude ( $z_a$ ) significantly degrades performance by exacerbating pointing errors. Conversely, increasing the transmitter depth ( $z_w$ ) from 10m to 30m surprisingly improves outage performance. This is attributed to a "beam-spreading trade-off": while deeper transmission increases attenuation, it significantly enlarges the beam footprint at the receiver, which mitigates wave-induced pointing errors. The geometric stabilization outweighs the attenuation penalty given the high sensitivity of the SiPM receiver.
Outage Floor: The derived outage probability curves flatten out at high wind speeds, confirming the existence of the TIR-induced outage floor. This floor is determined by the wind speed and the receiver's FoV.

Significance
The paper claims that this framework serves as an efficient tool for designing robust underwater data harvesting systems. By providing closed-form expressions, the model enables rapid performance evaluation without the computational burden of MC simulations. The identification of the TIR-induced outage floor offers critical insight for system designers, suggesting that reliability cannot be infinitely improved without addressing the fundamental physics of the water-air interface (e.g., through spatial diversity). The work aligns with ongoing efforts to model direct W2A channels, moving beyond simulation-heavy approaches to provide a mathematically tractable foundation for future system optimization.

The Big Picture: Sending a Flashlight from the Deep to the Sky

The Three Main Problems

The Surprising Discovery: Going Deeper Helps?

The Bottom Line

More like this