Randomness quantification in spontaneous emission

This paper establishes a comprehensive quantum information-theoretic framework to rigorously quantify intrinsic randomness in spontaneous emission-based QRNGs, demonstrating that while single-photon and temporal mode detection schemes are vulnerable to direct atom access, spatial mode and phase fluctuation schemes offer robust security against both direct and purified eavesdropping strategies.

The Gist

Imagine you are trying to generate a truly unpredictable number, like rolling a die that no one can cheat with. In the world of computers, we usually use "pseudo-random" generators, which are just complex math tricks. If you know the starting point and the rules, you can predict the future numbers. To get real randomness, we need to look at the quantum world, where nature itself is fundamentally unpredictable.

This paper by Chenxu Li, Shengfan Liu, and Xiongfeng Ma acts like a rigorous security audit for a specific type of quantum random number generator (QRNG) that uses spontaneous emission.

The Core Concept: The Atom and the Light

Think of an atom as a tiny, excited ball. When it relaxes, it drops a photon (a particle of light) into the universe. This is "spontaneous emission."

• The Paper's Insight: The randomness doesn't just come from the light; it comes from the entanglement (a deep, spooky connection) between the atom and the light it just dropped.
• The Analogy: Imagine the atom is a magician and the photon is a card they pull from a hat. Before the card is pulled, the magician and the card are in a superposition of all possibilities. The moment the card is pulled, the connection is broken, and a random result appears.

The Security Problem: The "Hacker" Scenarios

The authors ask a critical question: What if a hacker (Eve) is watching the magician? They define two types of hackers to test the security of different QRNG designs:

1. The "Inside Man" (Adversary I): This hacker has direct access to the atom itself. They can peek at the magician's hand before the card is pulled.
2. The "Ghost Observer" (Adversary II): This hacker cannot touch the atom, but they have a "ghost copy" (a purification) of everything the atom has ever emitted in the past. They are trying to guess the future based on old data.

The Four Methods: Which Ones Hold Up?

The paper tests four different ways to measure the light to generate numbers. Here is how they stack up against the hackers, using simple analogies:

1. Single-Photon Detection (The "Did it happen?" Check)

• How it works: You wait to see if a photon arrives in a specific time window. It's a simple "Yes" or "No."
• The Verdict: Vulnerable to the Inside Man.
• The Metaphor: If the hacker can touch the atom, they know exactly when the magician is about to drop the card. If the atom is in a "ready to drop" state, the hacker knows the answer is "Yes." The paper shows that if the hacker controls the atom, the randomness drops to zero.
• Against the Ghost: Surprisingly, it still holds some randomness against the Ghost Observer, even if the Ghost knows everything about the atom's past, because the act of dropping the card creates new randomness that the Ghost couldn't have predicted.

2. Temporal Mode (The "When did it happen?" Check)

• How it works: Instead of just asking "Did it happen?", you ask "Exactly when did it happen?" You divide time into tiny bins.
• The Verdict: Vulnerable to the Inside Man.
• The Metaphor: This is like the magician dropping a card at a specific second. If the hacker is holding the magician's hand, they know exactly which second the card will drop. The paper proves that if the hacker controls the atom, they can predict the exact time bin, rendering the randomness useless.
• Against the Ghost: Like the first method, it retains some security against the Ghost Observer, providing a lower bound of randomness.

3. Spatial Mode (The "Where did it land?" Check)

• How it works: You have an array of detectors all around the atom. You ask, "Which direction did the photon fly?"
• The Verdict: Secure against BOTH hackers.
• The Metaphor: Imagine the magician drops a card, but it flies in a superposition of all directions at once. When it hits a detector, it "collapses" into one specific direction.
• Why it's safe: The direction the photon flies is determined by the vacuum of space itself, not just the atom's internal state. Even if the hacker is holding the atom (Inside Man) or has a ghost copy of its past (Ghost Observer), they cannot predict which specific direction the photon will choose to fly because that choice is made by the interaction with the empty space around it. It's like the magician dropping a card that magically chooses a random wind to blow it in.

4. Phase Fluctuation (The "Wobble" Check)

• How it works: This looks at the "phase" (the timing of the wave) of a laser beam. The laser's phase wobbles randomly due to spontaneous emission.
• The Verdict: Secure against BOTH hackers.
• The Metaphor: Imagine a laser beam is a spinning top. Spontaneous emission is like tiny, invisible bugs bumping into the top, making it wobble randomly.
• Why it's safe: The wobble comes from the interaction between the laser and the vacuum (the empty space). Even if the hacker knows everything about the laser's atoms, they cannot predict the random bumps from the vacuum. As long as the hacker can't touch the vacuum interaction itself, the wobble remains truly random.

The Big Takeaway

The paper provides a mathematical "rulebook" to quantify exactly how much true randomness you can get from these systems.

• The Lesson: Not all quantum random number generators are created equal.
• If you use timing or simple detection, you must trust that no one is touching the atoms generating the light.
• If you use direction (spatial) or phase wobbles, the system is robust enough that even if a hacker has full access to the atoms, the randomness remains secure because it relies on the unpredictable nature of the vacuum itself.

The authors have built a framework that moves these devices from "we think this is random" (phenomenological) to "we can mathematically prove this is random and secure" (rigorous quantum information theory).

Technical Summary

Technical Summary: Randomness Quantification in Spontaneous Emission

Problem Statement
Quantum Random Number Generators (QRNGs) based on spontaneous emission are widely used due to their high speed and practical implementation. However, the quantification of intrinsic randomness in these systems has remained largely phenomenological. Existing analyses often lack rigorous adversarial models and fail to clearly characterize the role of quantum coherence in the generation process. Specifically, current models frequently assume that quantum phase fluctuations scale inversely with laser power or that photon statistics follow a Poisson distribution without deriving these from first principles. This absence of rigorous physical modeling compromises information-theoretic security, particularly regarding side-channel vulnerabilities where an adversary might access the atom ensemble underlying the spontaneous emission to predict outcomes via atom-field entanglement.

Methodology
The authors develop a comprehensive quantum information-theoretic framework to analyze intrinsic randomness in spontaneous emission processes from first principles.

1. Physical Model: The work utilizes the Wigner-Weisskopf theory of spontaneous emission, modeling the interaction between a two-level atom (system $A$) and the radiation field (system $R$) via a unitary evolution $U_{AR}$. The process is treated as the creation of an entangled state between the atom and the emitted field.
2. Adversary Models: Two distinct adversarial scenarios are defined based on the adversary's (Eve) access to the system:
   • Adversary I: Has direct (passive) access to the atom ensemble, allowing manipulation of the atomic state and measurement on the atom, but cannot manipulate the optical environment.
   • Adversary II: Does not have direct access to the atom but holds a purification of the atomic system (e.g., collected previous emissions or an ancilla system), representing a scenario of maximal information leakage without direct control.
3. Randomness Metric: Intrinsic randomness is quantified using the relative entropy of coherence, defined via conditional von Neumann entropy $S(A'|E)$. This metric measures the adversary's ignorance regarding the measurement outcomes. The analysis employs Positive-Operator-Valued Measures (POVMs) to model various detection schemes, utilizing Naimark extensions to relate generalized measurements to projective measurements on an extended Hilbert space.

Key Contributions and Results
The paper provides a rigorous quantification of extractable random bits for four specific QRNG schemes under the two adversary models. The results are summarized as follows:

• Single-Photon Detection (Temporal Mode):

  • Vulnerability: These schemes are vulnerable to Adversary I. If Eve controls the atom, she can predict the measurement outcomes because the randomness relies on the collapse of the atom-field superposition.
  • Security against Adversary II: Even if Eve holds the purification of the atom (maximal information leakage), the system guarantees a lower bound on intrinsic randomness. The unitary evolution generates fresh coherence that can be harvested, even if the initial atomic state is incoherent. The randomness is bounded by the atomic population ($\rho_{11}$) and the decay rate.
  • Temporal Mode Extension: Extending single-photon detection to temporal mode measurements (recording arrival times in time bins) increases the dimensionality of the entropy source. The analysis shows that while off-diagonal coherence terms affect the total randomness, a non-zero lower bound exists even when atomic coherence is zero.

• Spatial Mode Detection:

  • Robustness: QRNGs based on spatial mode detection (detecting the direction of photon emission) demonstrate security against both Adversary I and Adversary II.
  • Mechanism: The randomness arises from the quantum superposition over radiation modes created by spontaneous emission. Since these directional components are induced by vacuum fluctuations and not the atomic internal state, an adversary with access to the atom cannot predict the emission direction. The intrinsic randomness corresponds to the Shannon entropy of the spatial emission distribution.

• Quantum Phase Fluctuation:

  • Robustness: Similar to spatial mode schemes, phase-fluctuation-based QRNGs (using interferometers to measure relative phase) are secure against both adversary types.
  • Derivation: The authors derive the phenomenological "Gaussian white noise" model from first principles, showing that phase diffusion arises from the tracing out of vacuum modes coupled to the cavity. The security relies on the assumption that the adversary cannot access the environment during the cavity-vacuum interaction. If this assumption holds, the security is comparable to spatial mode schemes.

Significance and Claims
The paper claims to establish a unified perspective connecting spontaneous emission, quantum coherence, and randomness generation. Its primary significance lies in:

1. First-Principles Derivation: Moving beyond phenomenological assumptions (like Gaussian noise or Poisson statistics) to derive randomness quantification directly from the atom-field interaction Hamiltonian.
2. Security Hierarchy: Clarifying a trust hierarchy among QRNG protocols. The work demonstrates that while single-photon and temporal mode schemes require partial trust in the atomic ensemble (specifically that the adversary does not have direct access), spatial mode and phase-fluctuation schemes maintain intrinsic randomness even if the atomic subsystem is accessible to an adversary.
3. Quantitative Framework: Providing explicit formulas for the lower bounds of extractable random bits for different detection schemes, enabling proper post-processing and randomness extraction based on rigorous information-theoretic security rather than empirical noise characterization.

The authors emphasize that this framework is developed within a trusted-device scenario, distinct from device-independent approaches, focusing specifically on the physical modeling of the spontaneous emission source and the nature of the adversary's access to the atom versus the field.