🔢 mathematics

Application of uncertainty principles for decaying densities to the observability of the Schrödinger equation

This paper establishes observability inequalities for the Schrödinger equation in Euclidean space on measurable sets with decaying densities by leveraging quantitative uncertainty principles developed by Shubin, Vakilian, Wolff, and Kovrijkine.

Original authors: Kévin Le Balc'h, Jiaqi Yu

Published 2026-02-23

📖 5 min read🧠 Deep dive

CC0 1.0

Original authors: Kévin Le Balc'h, Jiaqi Yu

Original paper dedicated to the public domain under CC0 1.0 (http://creativecommons.org/publicdomain/zero/1.0/). ✨ 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 find a lost cat in a very large, foggy neighborhood. The cat is invisible, but you know it's moving around according to the laws of quantum physics (specifically, the Schrödinger equation). You have a few flashlights (sensors) placed in specific spots on the street.

The big question this paper answers is: How do you arrange your flashlights so that, no matter where the cat started, you can be 100% sure you've seen enough of it to figure out exactly where it began?

In the world of math, this is called Observability. If you can reconstruct the whole story from the parts you see, the system is "observable."

Here is the breakdown of what the authors, Kévin Le Balc'h and Jiaqi Yu, discovered, explained simply.

1. The Problem: The Cat is Spreading Out

In the quantum world, particles (like our cat) don't just stay in one spot; they spread out like a wave. If you only look at a tiny, fixed patch of the street, the wave might drift away, and you might miss it entirely.

For a long time, mathematicians knew that if your flashlights covered a "thick" enough area (meaning they weren't just thin lines or scattered dots, but had some substance everywhere), you could find the cat. But what if the neighborhood gets huge? What if the cat is more likely to be in the city center than in the distant suburbs?

2. The New Idea: "Decaying Density"

The authors introduce a clever new way to think about where to put the flashlights. Instead of requiring the lights to be everywhere equally, they suggest a smart gradient.

Imagine the neighborhood has a "density map."

In the city center (where the cat is likely to be): You need a lot of flashlights, packed tightly together.
In the distant suburbs (where the cat is unlikely to be): You can get away with fewer flashlights, spaced further apart.

The paper calls this a "Thick set with respect to a decaying density."

The Analogy: Think of it like a fishing net. Near the shore, the holes in the net are tiny (high density) so you catch small fish. Far out at sea, the holes get bigger (low density) because you expect fewer fish there. As long as the net is "thick" enough relative to how big the holes are, you will catch the fish.

3. The Secret Weapon: The Uncertainty Principle

How do they prove this works? They use a famous concept from physics called the Uncertainty Principle.

The Metaphor: Imagine you have a musical note. If you try to pinpoint exactly where the sound is coming from (location), you lose information about exactly what note it is (frequency/momentum). You can't know both perfectly at the same time.
The Application: The authors use a mathematical version of this principle. They prove that if your "net" (the observation area) is smartly designed to get thinner as you go further out, the wave function (the cat) cannot hide. It's mathematically impossible for the wave to be "invisible" in your smart net while still existing.

4. The Main Discoveries

The paper proves two main scenarios where this "smart net" works:

Scenario A: The "Good Enough" Net (Theorem 1.3 & 1.4)
If your net gets thinner at a specific rate as you go further out (specifically, if the spacing grows slower than the distance), you can find the cat.
- The Catch: If the net gets too thin too fast, you might miss the cat. But if it follows their specific recipe, you are guaranteed to find it, provided you wait long enough (or check at specific times).
Scenario B: The "Two-Check" System (Theorem 1.5)
What if you can't watch the cat all day? What if you only have two flashlights: one at 2:00 PM and one at 4:00 PM?
The authors show that if you have two different nets (one for the first time, one for the second) that follow their density rules, you can still figure out where the cat started. It's like taking two photos from different angles; even if the cat moved, the math connects the dots.

5. Why Does This Matter?

You might ask, "Who cares about finding a lost quantum cat?"

This math is crucial for Control Theory.

Engineering: It helps engineers design sensors for quantum computers or lasers. If you want to control a quantum system (like a qubit), you need to know exactly where it is. This paper tells you the most efficient way to place your sensors so you don't waste money covering empty space, while still guaranteeing you have total control.
Efficiency: It proves that you don't need a perfect, uniform grid of sensors. You can be smarter, placing more sensors where they matter most and fewer where they don't, saving resources.

Summary

The paper is a guide on how to be a smart observer. It tells us that we don't need to watch everything everywhere. Instead, we can use a "smart net" that gets looser the further out we go, relying on the fundamental laws of physics (uncertainty) to guarantee that we never lose track of the quantum particle.

In a nutshell: You don't need a net with holes the size of a pinhole everywhere. You just need a net where the holes get bigger just right as you move away from the center. If you do that, the math guarantees you'll catch the wave.

1. Problem Statement

The paper investigates the observability of the free Schrödinger equation in Euclidean space $\mathbb{R}^d$ :
$\begin{cases} i\partial_t u + \Delta u = 0 & \text{in } (0, +\infty) \times \mathbb{R}^d, \\ u(0, \cdot) = u_0 & \text{in } \mathbb{R}^d, \end{cases}$
where $u_0 \in L^2(\mathbb{R}^d)$ .

Observability refers to the ability to recover the full $L^2$ -norm of the initial state $u_0$ from measurements of the solution $u(t, x)$ restricted to a specific subset of space-time. The authors consider two distinct observation scenarios:

Continuous Observation: Observing the solution over a time interval $(0, T)$ on a spatial set $O \subset \mathbb{R}^d$ .
$\|u_0\|_{L^2}^2 \leq C \int_0^T \int_O |e^{it\Delta}u_0|^2 dx dt$
Two-Time Point Observation: Observing the solution at two distinct times $S$ and $T$ ( $T > S > 0$ ) on two spatial sets $O_1$ and $O_2$ .
$\|u_0\|_{L^2}^2 \leq C \left( \int_{O_1} |e^{iT\Delta}u_0|^2 dx + \int_{O_2} |e^{iS\Delta}u_0|^2 dx \right)$

The central challenge is to identify sufficient geometric conditions on the observation sets $O, O_1, O_2$ that guarantee these inequalities hold. While "thickness" (uniform density) is known to be necessary and sufficient in 1D, the multi-dimensional case ( $d \geq 2$ ) remains partially open, with existing results relying on strong conditions like the Geometric Control Condition (GCC) or finite complement measures.

2. Methodology

The authors employ a strategy combining Quantitative Uncertainty Principles with Resolvent Estimates.

A. Uncertainty Principles for Decaying Densities

The core analytical tool is a quantitative uncertainty principle established by Kovrijkine (inspired by Shubin, Vakilian, and Wolff). This principle relates the localization of a function $f$ and its Fourier transform $\hat{f}$ on sets with decaying densities.

Definition: A set $O$ is $(\gamma, \rho)$ -thick if $|O \cap B(x, \rho(x))| \geq \gamma |B(x, \rho(x))|$ for all $x$ , where $\rho(x)$ is a decaying density function (e.g., $\rho(x) \sim |x|^{-\alpha}$ ).
Key Inequality: If $O$ and $\Omega$ are thick with respect to densities $\rho_1$ and $\rho_2$ satisfying a specific reciprocal growth condition ( $C_1/\rho_1(C_2/\rho_2(t)) \geq t$ ), then:
$\|f\|_{L^2}^2 \leq C \left( \int_O |f|^2 + \int_\Omega |\hat{f}|^2 \right)$
This allows the authors to control the global norm of a function using its restriction to a set that becomes "thinner" at infinity, provided the density decays at a controlled rate.

B. Resolvent Estimates and Abstract Criteria

The authors translate the observability problem into spectral resolvent estimates using abstract results by Miller and Su-Sun-Yuan:

Miller's Criterion: Observability over $(0, T) \times O$ is equivalent to a resolvent estimate of the form $\|f\|^2 \leq M \|(-\Delta - \lambda)f\|^2 + m \|f\|_{L^2(O)}^2$ for all $\lambda \in \mathbb{R}$ .
High-Frequency Decay: For observability to hold for all $T > 0$ , the constant $M(\lambda)$ in the resolvent estimate must decay to zero as $\lambda \to \infty$ .
Proof Technique: The authors construct specific spectral projections (annuli in frequency space) and use the uncertainty principle to bound the $L^2$ -norm of functions supported in these annuli by their restriction to the observation set $O$ .

3. Key Contributions and Results

Result 1: Observability for Large Times (Theorem 1.3)

Condition: The observation set $O$ is $(\gamma, \rho)$ -thick where $\rho(x) = O(|x|^{-1})$ as $|x| \to \infty$ .
Conclusion: There exists a time $T_0 > 0$ such that the Schrödinger equation is observable in $(0, T) \times O$ for all $T \geq T_0$ .
Significance: This relaxes the requirement of a constant density (standard thickness) to a density that decays as $1/|x|$ . It proves that even if the observation set becomes sparse at infinity, observability is possible if the sparsity is not too severe.

Result 2: Observability for All Times (Theorem 1.4)

Condition: The observation set $O$ is $(\gamma, \rho)$ -thick where $\rho(x) = O(|x|^{-\alpha})$ with $\alpha > 1$ .
Conclusion: The equation is observable in $(0, T) \times O$ for every $T > 0$ .
Estimate: The observability constant behaves like $C e^{C/T^2}$ .
Significance: This is a sharp improvement over previous results. It shows that if the observation set is "thick enough" relative to a decay faster than $1/|x|$ , one can recover the initial state from arbitrarily short time intervals.

Result 3: Two-Time Point Observability (Theorem 1.5)

Condition: Two sets $O_1$ and $O_2$ are observed at times $T$ and $S$ . $O_1$ is $(\gamma, \rho_\alpha)$ -thick and $O_2$ is $(\gamma, \rho_{1/\alpha})$ -thick, where $\rho_\alpha(x) \sim |x|^{-\alpha}$ .
Conclusion: Observability holds for any $T > S > 0$ .
Significance: This generalizes the two-time point result from [WWZZ19] (which required finite complement sets) to sets with decaying densities, provided the densities of the two sets are reciprocal in their decay rates.

Extensions

The paper also extends these results to:

Fractional Schrödinger Equations: $i\partial_t u - (-\Delta)^s u = 0$ . The required decay rate of the density depends on $s$ (specifically $\alpha = 1/(2s-1)$ ).
Equations with Potentials: $i\partial_t u + \Delta u - V u = 0$ with $V \in L^\infty$ . The authors show that the high-frequency resolvent estimates hold even with a bounded potential, provided the observation set satisfies the thickness condition.

4. Significance and Comparison with Literature

Relaxation of Geometric Conditions: Previous sufficient conditions for multi-dimensional observability often required the observation set to satisfy the Geometric Control Condition (GCC) (every ray hits the set) or to have a finite complement ( $|O^c| < \infty$ $∣ O^{c} ∣ < \infty$ ).
- The authors demonstrate that these conditions are not necessary. They construct examples of sets that are $(\gamma, \rho)$ -thick with decaying $\rho$ but fail GCC and have infinite complements, yet still allow observability.
- Conversely, they show that sets satisfying GCC or finite complement conditions may fail the decaying thickness condition, highlighting that these are distinct geometric properties.
Sharpness: The results are shown to be sharp regarding the decay rate. If $\rho(x)$ decays faster than $|x|^{-1}$ (i.e., $|x|\rho(x) \to 0$ ), observability fails for the two-time point case.
Methodological Advancement: The paper successfully adapts Kovrijkine's uncertainty principles (originally for heat equations and band-limited functions) to the Schrödinger equation, bridging the gap between spectral theory and control theory in unbounded domains.

Conclusion

This work provides a new class of sufficient conditions for the observability of the Schrödinger equation in $\mathbb{R}^d$ . By introducing decaying densities, the authors significantly broaden the scope of admissible observation sets beyond those with constant density or finite complements. The results unify and extend previous findings, offering a precise characterization of how the "sparseness" of an observation set at infinity affects the ability to control or observe quantum systems.