⚛️ quantum physics

Markov Chain Model of Entanglement Setup in Noisy Dynamic LEO Satellite Networks

This paper proposes a comprehensive Markov chain model incorporating link storage age and physical distance to analyze and optimize quantum entanglement distribution in noisy dynamic LEO satellite networks, revealing critical trade-offs between request rates, fidelity, and satisfaction while validating the negligible impact of polarization rotation over short transmission distances.

Original authors: Yifan Gao, Alvin Valera, Winston K. G. Seah

Published 2026-03-16

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Yifan Gao, Alvin Valera, Winston K. G. Seah

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 send a secret, unbreakable message to a friend using Quantum Entanglement. Think of entanglement as a pair of "magic dice." If you roll a 6 on your die, your friend's die instantly shows a 6, no matter how far apart you are. This is the foundation of ultra-secure communication.

However, there's a catch: these magic dice are incredibly fragile. If you leave them sitting on a table, they start to "rot" (decohere) and lose their magic connection within a fraction of a second. Also, sending them through the air (free space) is like trying to throw a tiny marble from one moving train to another moving train while both are speeding around a curve.

This paper is a mathematical playbook for how to manage these fragile dice in a network of LEO (Low Earth Orbit) satellites. These satellites are like high-speed trains circling the Earth, passing each other quickly.

Here is the breakdown of their solution using simple analogies:

1. The Problem: The "Moving Target" Dilemma

In a normal fiber-optic cable (like the internet in your house), the path is fixed. But in space, satellites are constantly moving.

The Challenge: You can only create a "magic link" between two satellites if they are close enough (within about 40–50 km). If they are too far, the signal gets too weak.
The Timer: Once a link is made, it has a strict expiration date (about 0.2 seconds). If you don't use it by then, it rots and becomes useless.
The Traffic: Sometimes you have a message to send right now. Sometimes you have to wait for a message.

2. The Two Strategies: "Pre-cooking" vs. "Cooking to Order"

The authors used a Markov Chain (a fancy math tool that predicts the future based on the present) to compare two ways of handling these links:

Strategy A: The "Pre-cooked Meal" (Pre-generation)

How it works: The satellites constantly try to create magic links and store them in their "quantum fridge" (memory), waiting for a message to arrive.
The Good: When a message arrives, it's ready to go instantly! Very fast.
The Bad: If no message arrives quickly, the link in the fridge rots and gets thrown away. This is wastage.
The Verdict: Great if messages are rare, but you waste a lot of "magic dice" if you don't get many requests.

Strategy B: The "Cooking to Order" (On-demand)

How it works: The satellites wait until a message arrives. Then they start trying to create a magic link.
The Good: Zero waste. You never throw away a link because you only make it when you need it.
The Bad: You have to wait while they try to make the link. If they fail a few times, your wait gets longer.
The Verdict: Great if you have a steady stream of messages, but slow if you need instant answers.

3. The "Traffic Jam" Trade-off

The paper found a fascinating balance between speed and waste:

High Traffic (Many messages): If everyone is shouting for links at once, the "Pre-cooked" strategy works best because the links are used before they rot. However, if the satellites are too busy trying to make new links, they might miss some requests, lowering the success rate.
Low Traffic (Few messages): If messages are rare, "Cooking to Order" is better. "Pre-cooking" would just result in a fridge full of rotten links.

4. The "Spinning Top" Discovery (Polarization)

One of the coolest findings in the paper is about Polarization Rotation.

The Analogy: Imagine the magic dice are spinning tops. As they fly through space, the satellites' movement and mirrors can make the tops wobble or spin in the wrong direction, ruining the message.
The Finding: The authors proved that for short distances (40–50 km), this wobbling is so tiny it's basically zero. It's like worrying about a feather falling on a bowling ball.
Why it matters: Engineers can ignore this complex wobbling math for short hops, making the design of these satellite networks much simpler and cheaper.

5. The "Sweet Spot" for Distance

The paper calculated the perfect range for these satellites:

Too Close: Hard to find a partner satellite.
Too Far: The signal is too weak, and the "magic" fades before it arrives.
Just Right: 40 to 50 kilometers. At this distance, you get a good success rate, and the link stays fresh enough to use. They also found that the receiving telescope needs to be about the size of a large dinner plate (150mm) to catch enough "magic marbles."

Summary

This paper is like a traffic controller's manual for the Quantum Internet in space. It tells us:

Don't overstock: If you make too many links and don't use them, they rot.
Don't wait too long: If you wait to make a link, you might miss your window.
Keep it short: For now, keep the jumps between satellites short (40-50km) to avoid signal loss and complex math.
Ignore the wobble: For short jumps, you don't need to worry about the satellites spinning the signal out of alignment.

By understanding these rules, we can build a global quantum network that is fast, secure, and doesn't waste precious resources.

1. Problem Statement

Quantum entanglement routing is essential for scalable quantum communication, yet terrestrial fiber-based networks are limited by exponential photon loss to distances of ~100–200 km. Low Earth Orbit (LEO) satellites offer a solution via free-space channels with lower loss, but they introduce unique challenges:

Dynamic Topology: Satellites move at high velocities, causing rapidly changing inter-satellite distances and short communication windows (5–15 minutes).
Resource Constraints: Limited quantum memory capacity (assumed 1 qubit per satellite) and strict coherence time limits due to decoherence.
Noisy Environment: Free-space transmission suffers from geometric beam divergence, pointing errors, polarization rotation, and quantum memory decoherence.
Modeling Gap: Existing Markov chain models for quantum networks assume static topologies and fiber-based environments, failing to account for the dynamic distance variations and specific free-space loss mechanisms of LEO satellites.

The core problem is to analyze and optimize the trade-off between pre-generating entangled links (to reduce waiting time) versus on-demand generation (to avoid resource wastage from decoherence), while accounting for the probabilistic nature of link establishment and the degradation of fidelity over time and distance.

2. Methodology

The authors propose a comprehensive Markov Chain model specifically designed for dynamic LEO satellite networks.

A. System Modeling

State Space: A novel two-dimensional state space is introduced, defined by:
1. Physical Distance ( $d$ ): The Euclidean distance between the source and destination satellites.
2. Storage Age ( $i$ ): The time a link has been stored in quantum memory.
Physical Layer Analysis: The paper derives analytical expressions for link fidelity considering:
- Beam Diffraction: Gaussian beam spreading over distance.
- Pointing Errors: Modeled as a 2D Gaussian distribution affecting capture probability.
- Polarization Rotation: Caused by satellite motion and mirror jitter, modeled using rotation operators on the Bloch sphere.
- Decoherence: Modeled via an amplitude-damping channel in quantum memory, where fidelity decays exponentially ( $F(t) = F_0 e^{-2\Gamma t}$ ).
Protocol: The E91 protocol is used for link utilization, including quantum measurements, basis sifting, error estimation, and privacy amplification. The total time for generation and utilization is calculated based on light speed transmission and classical processing delays.

B. Strategies Analyzed

The model evaluates two distinct strategies under a "locked policy" (where the system rejects new requests while processing one):

Pre-generation Strategy: The system proactively generates links and stores them in memory.
- Pros: Reduces waiting time for incoming requests.
- Cons: Links may expire due to decoherence before use (resource wastage).
On-demand Strategy: The system only attempts to generate a link when a request arrives.
- Pros: No storage decoherence wastage; 100% link utilization efficiency.
- Cons: Higher waiting time due to generation latency.

C. Performance Metrics

The authors derive closed-form analytical expressions for four key metrics:

Request Satisfaction Rate: The probability that an arriving request is successfully served.
Average Waiting Time: The expected duration from request arrival to successful satisfaction.
Link Utilization Efficiency: The fraction of generated links that are successfully consumed (vs. wasted by expiration or failure).
Average Consumed Link Fidelity: The expected fidelity of links at the moment of utilization.

3. Key Contributions

First Dynamic LEO Model: This is the first Markov chain model to incorporate a two-dimensional state space (distance and age) specifically for dynamic LEO satellite quantum networks, addressing the limitations of static, fiber-based models.
Comprehensive Physical Modeling: The paper integrates complex free-space effects (beam divergence, pointing errors, polarization rotation, and decoherence) into the entanglement generation probability ( $p'$ ) and fidelity calculations.
Analytical Trade-off Analysis: It provides mathematical proofs and expressions revealing the fundamental trade-offs between request arrival rates, link fidelity, and satisfaction rates.
Design Guidelines: The study establishes practical engineering constraints for LEO quantum networks, including maximum transmission distances, required aperture sizes, and cutoff times.

4. Key Results & Findings

The evaluation uses realistic LEO parameters (e.g., Starlink-like orbits, Micius satellite specs) to derive the following insights:

Transmission Distance Limits:
- The maximum one-hop transmission distance is limited to 40–50 km under realistic error conditions (pointing error $\sigma_\delta \approx 0.5-1.0 \mu rad$ , QBER $\epsilon \approx 1-3\%$ ).
- At 40 km, the photon capture probability ranges from 56.74% to 65.27% depending on pointing errors.
- Initial fidelity ( $F'_0$ ) at 40 km ranges from 55.86% to 64.95%.
Decoherence and Cutoff Time:
- Due to rapid fidelity degradation, the cutoff time ( $T_{cutoff}$ ) for entangled links in LEO is extremely short, typically < 0.25 seconds (e.g., 0.244s at 40km with $\Gamma=0.5 s^{-1}$ ).
- Links stored longer than this threshold drop below the minimum fidelity threshold ( $F_{th}=0.5$ ) and must be discarded.
Polarization Rotation Impact:
- For short transmission distances (40–50 km), the impact of polarization rotation on maximum transmission distance is negligible (relative difference $\approx 10^{-7}\%$ ).
- Conclusion: Polarization rotation can be reasonably ignored in system design for short-hop LEO links, simplifying analysis.
Strategy Comparison (Pre-generation vs. On-demand):
- Low Request Rates ( $\lambda$ ): Pre-generation outperforms on-demand in satisfaction rate (e.g., +6.11% at $\lambda=0.05$ ) and waiting time (11–26% reduction) because links are ready. However, it suffers from low utilization efficiency (as low as 4.4% at 49km) due to decoherence wastage.
- High Request Rates: As $\lambda$ increases, pre-generation links are consumed faster, reducing decoherence effects. The satisfaction rates of both strategies converge, and pre-generation achieves higher link fidelity (up to 13.4% improvement at 40km) because links spend less time in memory.
- On-demand: Guarantees 100% link utilization efficiency but incurs higher waiting times and lower satisfaction rates at low traffic loads.
Parameter Sensitivity:
- Increasing the maximum generation attempts ( $G_{max}$ ) yields diminishing returns; increasing from 1 to 4 significantly improves satisfaction, but beyond 6, gains are minimal (<2%).
- A receiver aperture radius of 150 mm is recommended to achieve a maximum distance of ~50.78 km, while 120 mm is the minimum required for $F_{th}=0.5$ .

5. Significance

This work provides the theoretical foundation for designing and optimizing global-scale quantum communication networks using LEO satellites. By quantifying the trade-offs between latency, fidelity, and resource efficiency, the model enables:

Optimized Routing Algorithms: Designers can choose between pre-generation and on-demand strategies based on real-time network load and satellite dynamics.
Hardware Specifications: The results guide the selection of telescope aperture sizes, pointing accuracy requirements, and quantum memory coherence times.
Protocol Simplification: The finding that polarization rotation is negligible for short hops allows for simplified system designs, reducing computational and hardware overhead.

Ultimately, this model bridges the gap between theoretical quantum networking and the practical constraints of dynamic, noisy space environments, supporting the development of a future Quantum Internet.