Laplace-Carleson embeddings and infinity-norm admissibility

This paper provides a full characterization of the boundedness of Laplace–Carleson embeddings on $L^\infty$ and related Orlicz spaces in terms of Carleson intensity and weighted Berezin transforms, thereby establishing crucial criteria for the infinity-norm admissibility of control operators in linear diagonal semigroup systems.

The Gist

Imagine you are a traffic controller for a massive, complex city (the "system"). Cars (inputs) are coming in from all over, and your job is to make sure they flow through the city without causing a total gridlock or crashing the infrastructure.

In the world of mathematics and engineering, this "city" is a linear control system, and the "cars" are signals or inputs (like a voltage spike or a force applied to a bridge). The goal is to determine if the system can handle these inputs without blowing up.

This paper is about solving a very specific, tricky puzzle: What happens when the incoming traffic is "heavy" or "unpredictable"?

Here is the breakdown of the paper's ideas using everyday analogies:

1. The Two Types of Traffic (The Inputs)

Usually, mathematicians study systems where the traffic is "light" and follows a nice bell curve (called $L^2$ or $L^p$ spaces). It's like cars driving at a steady speed.

But in the real world, sometimes the traffic is extremely heavy and chaotic. Maybe a sudden storm hits, or a massive truck convoy arrives all at once. In math terms, this is $L^\infty$ (infinity-norm) admissibility. It means the input is "bounded" (it doesn't go to infinity), but it can be as rough and jagged as a saw blade.

The Problem: It is very hard to predict if a system can handle this "saw-blade" traffic. The old rules worked for smooth traffic, but they failed for the rough stuff.

2. The Magic Mirror (The Laplace Transform)

To analyze the traffic, the authors use a mathematical tool called the Laplace Transform. Think of this as a Magic Mirror.

• You look at the traffic in the real world (time domain).
• You look into the Magic Mirror, and suddenly, the chaotic traffic transforms into a pattern on a map (the complex plane).
• If the pattern on the map looks "safe" and contained, then the traffic in the real world is safe.

The paper focuses on a specific type of mirror called a Laplace–Carleson embedding. It's a way of checking if the "map" of the traffic fits inside a specific safety zone.

3. The "Intensity" of the Storm (Carleson Intensity)

The authors discovered a new way to measure the "intensity" of the storm.

Imagine you are checking a storm. You don't just look at the total rain; you look at how concentrated the rain is in specific areas.

• Old Rule: If the rain is spread out evenly, we are fine.
• New Rule (This Paper): Even if the rain is concentrated in specific "squares" on the map, we can still predict if the system will survive, provided the concentration isn't too extreme.

They introduced a formula (the Carleson Intensity) that acts like a storm meter. If the meter reads below a certain number, the system is safe, even with the roughest inputs.

4. The "Safety Net" (Orlicz Spaces)

Here is the most surprising discovery. The authors found that if a system can handle the roughest possible traffic ($L^\infty$), it automatically has a hidden safety net.

Think of it like this:

• If a bridge can hold a tank (the heaviest, roughest load), it can definitely hold a bicycle.
• But this paper says something cooler: If a bridge can hold a tank, it can also hold a very specific, weirdly shaped load that no one thought about before.

In math terms: If a system is "admissible" for $L^\infty$ (rough inputs), it is also admissible for a special class of "Orlicz spaces" ($L^\Phi$). These are like custom-tailored safety nets that fit perfectly between the smooth traffic and the rough traffic.

The authors proved that you don't just survive the rough traffic; you actually have a whole new category of "safe" traffic that you didn't know existed.

5. The "Diagonal" City (Diagonal Semigroups)

The paper focuses on systems that are "diagonal." Imagine a city where every street runs perfectly parallel to every other street, and they never cross. This makes the math easier to solve because you can analyze each street individually.

They proved that for these "parallel street" cities:

• If the system can handle the worst-case scenario ($L^\infty$), it implies it can handle a specific, slightly less rough scenario ($L^2$) if the city is built on a "Hilbert Space" (a very symmetrical, well-behaved type of city).
• However, if the city is built on a "Banach Space" (a more irregular, twisted city), the rules change, and you can't assume the system is safe just because it handles the rough stuff.

The Big Takeaway

Before this paper, engineers and mathematicians were stuck. They knew how to handle smooth inputs, and they knew how to handle some rough inputs, but the "worst-case" rough inputs were a mystery.

This paper provides the ultimate safety manual:

1. It gives a precise storm meter (Carleson intensity) to check if a system can handle the roughest possible inputs.
2. It reveals that if a system passes this test, it automatically gains a superpower: it can handle a whole new family of "weird" inputs (Orlicz spaces) that were previously unknown.
3. It solves a puzzle that has been open for years, helping engineers design better control systems for things like power grids, robotics, and structural engineering, ensuring they won't crash even when the inputs are chaotic.

In short: They figured out exactly how much "chaos" a system can take, and discovered that surviving the chaos makes the system stronger than anyone thought.

Technical Summary

1. Problem Statement

The paper addresses two interconnected mathematical problems:

1. Laplace–Carleson Embeddings: Characterizing the boundedness of the Laplace transform operator $L: Z \to L^q(\mathbb{C}_+, d\mu)$, where $Z$ is a function space on $(0, \infty)$, $\mathbb{C}_+$ is the open right half-plane, and $\mu$ is a positive regular Borel measure. While previous literature (e.g., [8, 9, 14]) had established results for $Z = L^p$ with $1 \le p < \infty$, the case $Z = L^\infty$ (corresponding to bounded inputs) remained largely open and technically difficult.
2. Admissibility of Control Operators: In the context of linear control systems $\dot{x}(t) = Ax(t) + Bu(t)$, determining when a control operator $B$ is admissible. Specifically, the authors focus on $L^\infty$-admissibility (bounded inputs) and its relationship to admissibility in Orlicz spaces ($L^\Phi$) and $L^p$ spaces for diagonal semigroups.

The core challenge is that $L^\infty$-admissibility is the most physically relevant case for bounded inputs but the hardest to analyze analytically. The authors aim to provide necessary and sufficient conditions for this boundedness and explore the implications for system stability.

2. Methodology

The authors employ a combination of harmonic analysis, functional analysis, and control theory:

• Carleson Intensity and Dyadic Decomposition: They introduce the concept of $\alpha$-Carleson intensity, $C_\alpha[\mu]$, defined via Carleson squares $Q_I$. A key technique involves decomposing the measure $\mu$ into dyadic strips $S_n = \{x+iy : 2^n \le x < 2^{n+1}\}$ and analyzing the measure $\mu_n$ restricted to these strips.
• Berezin Transforms: The authors utilize weighted Berezin transforms ($B_\alpha\mu$ and $\tilde{B}_\alpha\mu$) to characterize the measure $\mu$. They establish equivalences between the finiteness of integrals involving these transforms and the boundedness of the embedding.
• Orlicz Spaces and N-functions: The paper extensively uses Orlicz spaces $L^\Phi$ defined by Young functions $\Phi$. They focus on N-functions (a specific class of Young functions satisfying growth conditions at 0 and $\infty$) to construct intermediate spaces between $L^\infty$ and $L^p$.
• Test Functions and Duality: To prove necessity, they construct specific test functions (often involving characteristic functions and oscillatory terms like $e^{ict}$) to derive lower bounds on the operator norm. For sufficiency, they utilize the Hausdorff–Young theorem and classical Carleson embedding theorems for Hardy spaces.
• Diagonal Semigroups: In the control theory section, they restrict attention to systems where the generator $A$ is diagonal with respect to a $q$-Riesz basis, allowing the admissibility problem to be mapped directly to the Laplace–Carleson embedding problem via the measure $\mu = \sum |b_k|^q \delta_{-\lambda_k}$.

3. Key Contributions and Results

A. Characterization of $L^\infty \to L^q$ Embeddings

The paper provides a full characterization of the boundedness of $L: L^\infty(0, \infty) \to L^q(\mathbb{C}_+, d\mu)$ for $q \ge 2$.

• Theorem 2.3: The embedding is bounded if and only if the sum of Carleson intensities over dyadic strips is finite:
  $$\sum_{n \in \mathbb{Z}} C_q[\mu_n] < \infty$$
  This condition is shown to be equivalent to integral conditions involving the Berezin transform:
  $$\int_0^\infty t^{1-q} \sup_{\text{Re } z = t} \tilde{B}_\alpha\mu(z) \, dt < \infty$$
  (under specific constraints on $\alpha$ and $q$).
• Contrast with $L^p$: Unlike the $L^p$ case where a single Carleson condition on the whole measure often suffices, the $L^\infty$ case requires the summability of intensities across different scales (dyadic strips).

B. Implications for Orlicz Spaces

A major theoretical contribution is the link between $L^\infty$ and Orlicz spaces:

• Theorem 2.10: If the embedding $L: L^\infty \to L^q$ is bounded, there exists an N-function $\Phi$ such that the embedding $L: L^\Phi \to L^q$ is also bounded.
• This implies that $L^\infty$-admissibility is "stronger" than $L^\Phi$-admissibility for a specific class of Orlicz spaces, effectively bridging the gap between bounded inputs and $L^p$ inputs.

C. Finite Time and Specific Orlicz Spaces

• Finite Time: The results are extended to functions supported on $(0, \tau_0)$. The authors show that the boundedness on a finite interval implies boundedness on any smaller interval, and the condition remains the summability of Carleson intensities (Theorem 2.13).
• Exponential Growth: For the specific Orlicz space defined by $\Phi(t) = e^t - t - 1$, the authors derive a specific characterization involving a weighted sum of Carleson intensities with a factor of $n^2$ (Theorem 2.16 and 2.18).

D. Admissibility of Control Operators

The abstract results are applied to linear control systems:

• Diagonal Semigroups: For a diagonal semigroup generated by $A$ with respect to a $q$-Riesz basis ($q \ge 2$), $L^\infty$-admissibility of $B$ is equivalent to the condition $\sum C_q[\mu_n] < \infty$ (Theorem 3.6).
• Implication for Stability: The authors prove that for diagonal semigroups, $L^\infty$-admissibility implies $L^\Phi$-admissibility for some N-function $\Phi$ (Corollary 3.7).
• Left-Invertible Semigroups: For left-invertible semigroups on Hilbert spaces, they show that $L^\infty$-admissibility is equivalent to $L^2$-admissibility (Theorem 3.3), generalizing previous results on wave equations.
• Zero-Class Admissibility: The paper establishes that $L^\infty$-admissible operators for diagonal systems are automatically zero-class admissible (the norm of the input-to-state map vanishes as the time horizon $t_0 \to 0$).

4. Significance

• Resolution of Open Problems: The paper answers questions that have been implicitly open for several years regarding the characterization of $L^\infty$-admissibility, which is crucial for systems with bounded inputs.
• Bridging Spaces: It establishes a rigorous hierarchy showing that $L^\infty$-admissibility implies admissibility in specific Orlicz spaces, providing a new tool for analyzing input-to-state stability (ISS).
• New Characterizations: The introduction of the summability of dyadic Carleson intensities as the necessary and sufficient condition for $L^\infty$ embeddings offers a precise analytical tool that was previously unavailable.
• Applications: The results are directly applicable to the analysis of boundary control systems and infinite-dimensional systems where bounded inputs are the norm, offering new criteria for verifying system well-posedness and stability.

In summary, the paper successfully extends the theory of Carleson embeddings to the $L^\infty$ and Orlicz space settings, providing a complete characterization for diagonal systems and revealing deep connections between bounded input admissibility and input-to-state stability.