⚛️ quantum physics

Asymptotic Scaling of Precision Limits in Continuous Gaussian Quantum Metrology

This paper establishes a general theoretical framework for continuous Gaussian quantum metrology, deriving analytical expressions for asymptotic precision limits to show that while global quantum Fisher information can exhibit quadratic or even exponential scaling under specific conditions like nonreciprocity, the accessible environmental information is fundamentally constrained to linear scaling, revealing a critical distinction between stored and radiated quantum information.

Original authors: Kazuki Yokomizo, Aashish A. Clerk, Yuto Ashida

Published 2026-08-28

📖 6 min read🧠 Deep dive

CC BY 4.0

Original authors: Kazuki Yokomizo, Aashish A. Clerk, Yuto Ashida

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 quest to measure the world with perfect precision is a driving force behind much of modern science and technology. From detecting the faint ripples of gravity from colliding black holes to imaging the delicate machinery inside a living cell, the ability to sense tiny changes defines the frontier of discovery. For decades, scientists have known that the laws of quantum mechanics offer a path to surpass the limits of classical measurement. By using the strange properties of quantum particles, such as their ability to exist in multiple states at once or to be linked across distances, researchers can theoretically build sensors that are far more sensitive than anything made of ordinary matter. However, most of our understanding of these quantum limits comes from studying systems that are measured just once, like a snapshot of a particle's position.

A different, more dynamic approach involves watching a quantum system continuously over time. In this scenario, the sensor is an open system that constantly interacts with its surroundings, leaking information into the environment like a stream of light or particles escaping a container. The challenge here is to understand how much useful data can be extracted from this continuous stream. Does the information simply accumulate in a predictable way, or does the complex interplay between the sensor and its environment create new, unexpected possibilities? This question is particularly important for systems made of bosons, a type of particle that includes photons of light and vibrations in solid materials, which are the building blocks of many advanced quantum technologies.

In a recent study, researchers developed a comprehensive theoretical framework to answer these questions for continuous sensing with bosonic systems. They focused on a specific type of measurement where the system is monitored in a way that preserves its quantum nature, allowing them to calculate exactly how much information is stored in the combined system and environment versus how much is actually carried away by the escaping radiation. Their work reveals that the rules governing these sensors are far more nuanced than previously thought, depending heavily on whether the system loses energy to its surroundings or retains it, and on how the different parts of the sensor are connected to one another.

The researchers began by distinguishing between two types of information. The first is the total information encoded in the joint state of the sensor and the environment, which represents the ultimate limit of what could be known if one could measure everything perfectly. The second is the environmental information, which is the maximum amount of data that can be practically retrieved by simply watching the radiation or particles leaking out of the system. In many practical situations, scientists can only access this leaked information. The study found that in systems where the sensor loses energy and settles into a stable state, these two amounts of information eventually grow at the same rate. This means that in such dissipative systems, the information radiated away faithfully reflects the total information generated, leaving no hidden data trapped inside the sensor.

However, the picture changes dramatically when the system does not lose energy to the environment. In these zero-damping scenarios, the sensor can retain coherent excitations, meaning it holds onto information that is not immediately radiated away. Here, the total information stored in the system and environment grows faster than the information that leaks out. The researchers showed that this gap between what is stored and what is accessible depends on the specific way the sensor is built and how it is measured. They identified that while the total information can be enhanced by connecting many parts of the sensor together, the information that escapes to the environment follows different rules, often scaling more slowly.

To test these ideas, the team analyzed several concrete physical setups, including arrays of coupled optical cavities and arrays of trapped particles. They found that if each part of the sensor is measured independently, the precision improves linearly with the number of parts, which is a standard result. But if the sensor is designed so that all parts are measured collectively, the precision can improve quadratically, meaning that doubling the number of parts yields four times the sensitivity. This quadratic scaling represents a significant quantum advantage, allowing for much more precise measurements with the same amount of resources.

Perhaps the most striking discovery involved a setup where the sensor was designed to be nonreciprocal, meaning that energy or information flows preferentially in one direction. In this configuration, the researchers found that the total information stored within the system could grow exponentially with the number of parts, a massive leap compared to the linear or quadratic growth seen in other setups. This exponential boost is driven by a phenomenon known as the non-Hermitian skin effect, where quantum states become localized at the edges of the system, amplifying the response to external signals. However, this exponential enhancement does not appear in the information that leaks out to the environment. The radiation field still carries information that scales only linearly with the number of parts. This reveals a fundamental distinction: the system can harbor a vast amount of hidden information that is not accessible through simple monitoring of the output.

The study also addressed the role of energy as a resource. They demonstrated that while increasing the number of particles or modes in the sensor can boost precision, the growth is ultimately limited by the energy available. Even in the most favorable scenarios, the precision scales at most linearly with the energy resource, meaning that simply pumping more energy into the system does not lead to infinite sensitivity. The researchers also explored what happens when the total strength of the measurement is kept constant regardless of the number of parts. In dissipative systems, this normalization reduces the quadratic advantage back to a linear one, but in the zero-damping case, the quadratic scaling of the total information persists, highlighting the unique potential of systems that do not lose energy.

These findings provide a clear map of the trade-offs involved in continuous quantum sensing. They show that while collective measurements and nonreciprocal designs can unlock powerful scaling behaviors, the information that is actually radiated into the environment may not always reflect the full potential of the system. For scientists designing future sensors, this means that the choice of how to connect the sensor to the outside world is just as critical as the internal design of the sensor itself. The work suggests that by carefully engineering the flow of information and energy, it may be possible to build sensors that approach the fundamental limits of precision, offering new capabilities for detecting the faintest signals in the universe. The researchers note that while their results are based on theoretical models and simulations, the physical setups they described, such as arrays of trapped particles or coupled cavities, are within reach of current experimental capabilities, pointing toward a future where these theoretical advantages can be realized in the laboratory.

Technical Summary: Asymptotic Scaling of Precision Limits in Continuous Gaussian Quantum Metrology

Problem Statement
Continuous quantum metrology offers a pathway to high-precision sensing by progressively extracting information from radiation quanta emitted by an open quantum system. While significant progress has been made in discrete-variable systems (e.g., qubits) and interferometric settings, a comprehensive theoretical understanding of precision limits in continuous sensing with multimode bosonic systems remains lacking. Specifically, there is a need to characterize how the Quantum Fisher Information (QFI)—the metric for estimation precision—scales with physical resources (such as the number of modes $M$ and bosonic excitations $\bar{n}$ ) in the long-time limit. Furthermore, the relationship between the total information encoded in the joint system-environment state (Global QFI, $I_G$ ) and the information accessible solely from the emitted radiation (Environmental QFI, $I_E$ ) requires clarification, particularly regarding whether these quantities scale identically or diverge under different coupling regimes.

Methodology
The authors develop a general theoretical framework for quantum metrology involving multimode free bosons subjected to continuous Gaussian measurements. The analysis focuses on linear dynamics where the system Hamiltonian is at most quadratic in quadrature operators, and jump operators are linear combinations thereof. This ensures the system state remains a bosonic Gaussian state, fully characterized by its first and second statistical moments (mean vector and covariance matrix).

Key methodological steps include:

Formalism: Utilizing the generalized Lindblad equation to track a pseudo-reduced density matrix $\hat{\mu}_{\theta_1, \theta_2}(t)$ , which evolves solely within the system Hilbert space. This allows for the calculation of $I_G$ without explicitly tracking the growing environmental Hilbert space.
QFI Derivation: Deriving analytical expressions for the asymptotic growth rates of $I_G$ and $I_E$ . The global QFI is linked to the trace of the pseudo-reduced density matrix, while the environmental QFI is linked to its trace norm. The difference between them is shown to depend on the imaginary part of the mean vector of the pseudo-state.
Coupling Classes: The study categorizes system-environment couplings into two distinct classes:
- Dissipative Couplings: Systems evolving towards a unique steady state (e.g., coupled cavity arrays with photon leakage).
- Zero-Damping Couplings: Systems exhibiting purely oscillatory dynamics with unbounded energy growth (e.g., trapped particle arrays with backaction-evading measurements).
Resource Analysis: Defining optimized QFI rates ( $\dot{I}^*_{G,st}$ and $\dot{I}^*_{E,st}$ ) by maximizing over linear signal profiles. The authors derive bounds for these rates in terms of the number of modes $M$ and the bosonic excitation number (or its growth rate).

Key Contributions and Results

Asymptotic Scaling Laws:
The paper establishes that in the long-time limit, both $I_G$ and $I_E$ grow linearly with time ( $t$ ). The focus is on the asymptotic growth rates ( $\dot{I}_{st}$ ).
- Dissipative Regime: For systems with dissipative couplings, the asymptotic growth rates of the global and environmental QFIs coincide ( $\dot{I}_{G,st} = \dot{I}_{E,st}$ ). The optimized rates are bounded by $r_1 \bar{n}_{st} M \leq \dot{I}^* \leq r_2 \bar{n}_{st} M^2$ , where $\bar{n}_{st}$ is the steady-state excitation per mode. This indicates that while quadratic scaling with mode number $M$ is possible (via global coupling), the scaling with energy resources is at most linear.
- Zero-Damping Regime: For systems without damping, the rates generally separate ( $\dot{I}_{G,st} > \dot{I}_{E,st}$ ). The global QFI rate is bounded by $r_3 \dot{\bar{n}}_{st} M \leq \dot{I}^*_{G,st} \leq r_4 \dot{\bar{n}}_{st} M^2$ , where $\dot{\bar{n}}_{st}$ is the excitation growth rate. Crucially, the environmental QFI rate obeys a different bound: $r_5 M \leq \dot{I}^*_{E,st} \leq \min(r_6, r_7 \dot{\bar{n}}_{st}) M^2$ . This reveals that in zero-damping systems, the information radiated into the environment does not necessarily scale with the internal excitation energy in the same way as the total information.
Specific Setup Analyses:
- Local vs. Global Coupling: In local setups (independent monitoring of modes), both QFIs scale linearly with $M$ ( $I \propto Mt$ ). In global setups (collective monitoring), both QFIs can achieve quadratic scaling with $M$ ( $I \propto M^2t$ ) in both dissipative and zero-damping regimes.
- Nonreciprocal Setup: The authors analyze a nonreciprocal local setup where non-Hermitian skin effects induce unidirectional energy flow. This leads to an exponential scaling of the bosonic excitation rate ( $\dot{\bar{n}}_{st} \propto e^{O(M)}$ ) and, consequently, an exponentially enhanced global QFI ( $\dot{I}_{G,st} \propto e^{O(M)}$ ). However, the environmental QFI retains only linear scaling with $M$ ( $\dot{I}_{E,st} \propto Mt$ ). This demonstrates a fundamental distinction: the exponential enhancement is stored in the system-environment correlations and internal system dynamics but is not reflected in the information radiated to the environment.
Fixed Coupling Strength:
The paper also examines the scaling under a fixed total coupling strength convention. In dissipative systems, the quadratic scaling of the global QFI reduces to linear scaling when the collective channel is normalized. In zero-damping systems, the global QFI retains quadratic scaling with $M$ even after normalization, while the environmental QFI shifts from quadratic to linear scaling.

Significance and Claims
The authors claim that this work establishes a foundational understanding of resource trade-offs and scaling behaviors in continuous bosonic sensing. The primary significance lies in:

Bridging the Gap: Providing a rigorous analytical framework for continuous metrology in multimode bosonic systems, a domain previously less explored compared to discrete-variable or interferometric settings.
Fundamental Distinction: Highlighting a qualitative difference between the information stored in the joint system-environment state and the information radiated into the environment, particularly in zero-damping and nonreciprocal scenarios. The finding that nonreciprocal setups can yield exponentially enhanced global QFI without a corresponding enhancement in environmental QFI challenges the assumption that internal resource amplification always translates to accessible measurement precision.
Resource Bounds: Defining precise bounds on precision limits, showing that while quadratic scaling with mode number is attainable, the scaling with energy resources is fundamentally limited to linear growth in dissipative systems, and the relationship between internal energy and accessible information is more complex in zero-damping systems.
Experimental Relevance: The theoretical models are motivated by and applicable to concrete experimental platforms, including coupled cavity arrays and trapped particle arrays (e.g., levitated nanoparticles or ultracold atoms), suggesting that these scaling behaviors are physically realizable.

The paper concludes that the interplay between coherent dynamics and measurement backaction in bosonic many-particle systems creates a rich landscape for continuous quantum metrology, offering unique scaling behaviors and resource trade-offs that differ fundamentally from single-shot interferometric counterparts.

Technical Summary: Asymptotic Scaling of Precision Limits in Continuous Gaussian Quantum Metrology

More like this