Stabilization of Time-Varying Perturbed Quantum Systems via Reduced Filters
Weichao Liang, Daoyi Dong · 2025
Read Stabilization of Time-Varying Perturbed Quantum Systems via Reduced Filters by Weichao Liang, Daoyi Dong (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2511-07949.txt.
Introduction
Weichao Liang and Daoyi Dong propose a lighter filter for stabilizing uncertain quantum systems through feedback. Real quantum devices face unknown starting states, imprecise parameters, and extra environmental noise. Full filters track a number of variables that grows with the square of the system size, which gets costly fast.
The reduced filter tracks only diagonal density matrix entries in a chosen measurement basis, a count growing just linearly with size. That cut makes real time computation and feedback much simpler. Analysis still gives global exponential stability of the target subspace, plus robustness against time varying changes and dissipative effects.
Catalog record
| CREATOR | Weichao Liang, Daoyi Dong |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 76 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2511-07949
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Stabilization of Time-Varying Perturbed Quantum Systems via Reduced Filters Weichao Liang, Daoyi Dong Abstract In practical applications, quantum systems are inevitably subject to significant uncertainties, including unknown initial states, imprecise physical parameters, and unmodeled environmental noise, all of which pose major challenges to robust quantum feedback control. This paper proposes a feedback stabilization strategy based on a reduced quantum filter that achieves robustness against time-varying Hamiltonian perturbations and additional dissipative effects, without requiring prior knowledge of the initial state or exact system parameters. The proposed filter estimates only 𝒪 ( N ) \mathcal{O}(N) real variables corresponding to the diagonal elements of the system density matrix in a quantum non-demolition basis in contrast to the 𝒪 ( N 2 ) \mathcal{O}(N^{2}) variables required by a full stochastic master equation, where N N is the Hilbert space dimension. This dimensionality reduction substantially simplifies real-time computation and feedback implementation while preserving both convergence and robustness guarantees. Rigorous analysis further establishes global exponential stability of the target subspace. The results provide a scalable framework for robust and efficient measurement-based feedback control applicable to high-dimensional perturbed open quantum systems. 1 INTRODUCTION The theory of open quantum systems [ 11 ] , describing systems interacting with an external environment, has profoundly impacted quantum information science [ 31 ] . Such interactions inevitably induce quantum dissipation and decoherence, leading to information loss, one of the most critical challenges in quantum control and quantum computation. Developing feedback control strategies to mitigate decoherence is crucial for the scalability and reliability of quantum technologies, including quantum computing, quantum chemistry, and quantum information processing [ 13 , 1 ] . A widely adopted framework for analyzing quantum control in the presence of continuous measurement is based on Stochastic Master Equations (SMEs) [ 6 , 10 , 3 ] . In this framework, an open quantum system interacts with its environment and a probe system, where the probe system is continuously monitored. SME-based feedback control has been a key enabler for various quantum technologies [ 46 ] , leading to significant advances in various domains [ 46 ] : Quantum state protection , extending coherence lifetimes [ 45 ] and protecting macroscopic Schrödinger cat states [ 32 ] ; Quantum error suppression and correction , mitigating decoherence and enabling real-time error suppression [ 27 ] ; Quantum simulation , stabilizing Bose-Einstein condensates to improve quantum simulations of condensed matter systems [ 38 ] ; Quantum transport in nano-structures : offering new opportunities in quantum device engineering; Quantum sensing and metrology , increasing sensor sensitivity [ 14 ] ; Entanglement distribution , enabling robust remote entanglement generation and stabilization, critical for quantum communication and distributed computing [ 35 ] . Recent research has also explored its role in connecting gravity and quantum matter in the so-called Newtonian limit [ 33 ] . Despite these successes, two major challenges remain for the practical implementation of SME-based feedback control: • Computational complexity of quantum filtering : Traditional full-state quantum filters track \mathcal{O}(N^{2}) parameters for an N -level quantum system, making real-time feedback control infeasible for large quantum systems [ 42 ] . Several model reduction methods have been proposed to address this challenge [ 44 , 30 , 15 , 16 , 2 , 17 ] , but their applicability remains restricted. • Robustness to uncertainties : Unknown initial states, detector inefficiencies, model inaccuracies, and time-varying parameters (e.g., coupling strengths, measurement efficiencies) reduce the effectiveness of a control strategy. Existing robust methods often assume structured uncertainties or full-state estimation [ 25 , 24 , 23 , 18 ] , limiting practical applicability. These issues are also closely related to non-Markovian effects [ 11 , 4 , 5 ] . Earlier stabilization approaches have largely relied on engineering Lindbladian dynamics [ 41 , 29 , 39 , 8 ] , focusing on deterministic dissipator design while paying less attention to measurement back-action, i.e., the diffusion terms in the SME. Although valuable structural conditions were derived, such methods often assume precise knowledge of the system state and parameters. In contrast, our earlier work [ 20 , 23 , 24 , 25 ] has shown that directly leveraging measurement back-action makes it possible to design feedback laws that guarantee exponential stabilization, while remaining robust against both unknown initial states and parameter uncertainties within admissible ranges. This highlights a clear advantage of measurement-based feedback strategies over purely Lindbladian designs. Specifically, in [ 26 ] , we first introduced a reduced-filter-based approach for stabilizing perturbed systems with time-invariant coefficients. This work extends our prior results by developing a rigorous framework for robust feedback stabilization of time-varying perturbed quantum systems using a reduced-order filter. This approach addresses both the computational challenges and robustness issues that arise when dealing with large quantum systems subject to time-varying perturbations. Our key idea is to construct a reduced filter that tracks only \mathcal{O}(N) real parameters, corresponding to the diagonal entries in a Quantum Non-Demolition (QND) basis, rather than the full \mathcal{O}(N^{2}) -dimensional state. This substantially reduces computational costs and enables scalable real-time feedback control for large quantum systems. Our main contributions are summarized as follows: 1. Reduced filter construction : We rigorously establish the existence and well-posedness of the reduced filter and derive sufficient conditions for its systematic construction in an intuitive manner (Theorem 4.1 ). 2. Robust exponential stabilization : We prove that the reduced-filter-based feedback globally exponentially stabilizes the quantum system to the desired target subspace, under time-varying model perturbations. Unlike prior methods, our approach eliminates the need for full-state estimation or precise system parameters. Additionally, we relax constraints on the feedback controller compared to our previous work [ 23 , 25 ] (Theorem 4.10 ). Beyond rigorous proofs, we provide intuitive insights into the reduced filter’s construction and the rationale behind the proposed conditions, ensuring robust stabilization. This paper is organized as follows: In Section 2 , we introduce the stochastic dynamical model of open quantum systems under continuous-time measurements and present the control problem. Section 3 discusses the large-time behavior of uncontrolled systems, focusing on exponential quantum state reduction. In Section 4 , we present the reduced-filter-based feedback design and explore its robustness for stabilizing the open quantum system. Section 5 provides a numerical example of a three-level system. Finally, Section 6 concludes the paper and discusses future directions for research. Notation. The imaginary unit is denoted by \mathfrak{i} . For X\in\mathcal{B}(\mathcal{H}) , the adjoint is X^{*} , where \mathcal{B}(\mathcal{H}) denotes the space of all linear operators on \mathcal{H} , and \mathbf{I} denotes the identity operator. The set of Hermitian operators is \mathcal{B}_{*}(\mathcal{H}):=\{X\in\mathcal{B}(\mathcal{H})\mid X=X^{*}\}. The trace of X\in\mathcal{B}(\mathcal{H}) is \mathrm{Tr}(X) , and the Hilbert–Schmidt norm is \|X\|:=\mathrm{Tr}(XX^{*})^{1/2}. For v\in\mathcal{H} , \|v\| denotes the standard vector norm. Define L^{\infty}([0,\infty),\mathcal{B}(\mathcal{H})):=\{X:[0,\infty)\to\mathcal{B}(\mathcal{H})\mid\text{each entry is measurable and }\sup_{t\geq 0}\|X(t)\|<\infty\} . The commutator of X,Y\in\mathcal{B}(\mathcal{H}) is [X,Y]:=XY-YX . For x\in\mathbb{C} , \Re\{x\} denotes the real part. The Hölder space of functions from A to B is denoted \mathcal{C}^{1,\alpha}(A,B) for \alpha\in(0,1] . For a positive integer m , we write [m]:=\{1,\dots,m\} . If \mathcal{H}=\mathcal{H}_{S}\oplus\mathcal{H}_{R} and X\in\mathcal{B}(\mathcal{H}) , then in a basis adapted to this decomposition X has block form X=\begin{bmatrix}X_{S}&X_{P}\\ X_{Q}&X_{R}\end{bmatrix}, where X_{S},X_{R},X_{P} and X_{Q} are matrices representing operators from \mathcal{H}_{S} to \mathcal{H}_{S} , from \mathcal{H}_{R}