Feedback control of linear quantum optical systems

Hu Zhang, Guofeng Zhang, Daoyi Dong, Bo Huang, Joseph H. W. Lee2011 · DOI

TL;DR

This paper introduces a novel quadrature representation for linear quantum optical systems that explicitly incorporates phase modulation. This new framework allows for a more systematic approach to designing feedback controllers, leading to improved performance in quantum optical systems, as demonstrated by a significant reduction in a key performance metric.

What Problem Does This Paper Address?

Controlling quantum systems is notoriously challenging due to their inherent fragility and the probabilistic nature of quantum mechanics. Feedback control offers a powerful way to stabilize and manipulate these systems, but designing effective controllers for quantum optical systems, which are often described by complex linear dynamics and subject to quantum noise, remains a significant hurdle. Existing methods often struggle to systematically incorporate crucial elements like phase modulation and squeezing, which are vital for enhancing control performance. This paper aims to bridge this gap by developing a new mathematical framework that simplifies the analysis and design of feedback controllers for these systems, specifically focusing on improving Linear Quadratic Gaussian (LQG) control performance.

Methods

The core contribution of this paper lies in the introduction of a new "quadrature representation" for linear quantum optical systems. Unlike previous approaches that primarily used annihilation and creation operators, this new representation explicitly incorporates phase modulation. This is achieved by defining a set of real-valued operators and matrices that capture the system dynamics, including the effects of phase modulators and ideal squeezers.

Key methodological steps include:

  • New Quadrature Representation: Developing a mathematical framework that transforms the system description from annihilation/creation operators to real-valued quadratures, explicitly accounting for phase modulation. This representation is shown to be more amenable to control design.
  • Controller Parameterization: Leveraging the new representation to parameterize the controller, including its internal dynamics, squeezing components, and direct coupling mechanisms. This parameterization ensures physical realizability by construction.
  • Two-Stage Optimization: Proposing a two-stage optimization algorithm for controller design. The first stage uses a Genetic Algorithm (GA) for a broad global search to find an initial set of controller parameters. The second stage employs a local search method to refine these parameters and find a near-optimal solution, leveraging the continuous differentiability of the cost function and constraints.
  • LQG Control Formulation: Re-investigating coherent quantum LQG control within this new framework, aiming to minimize a quadratic performance index.

Datasets

Not discussed in the paper. The paper focuses on theoretical development and numerical examples rather than training on specific datasets.

Key Insights and Ideas

  • Phase Modulation is Key: The paper highlights that explicitly incorporating phase modulation into the system representation is crucial for achieving superior control performance. This is a significant departure from previous approaches that might have treated it as a secondary effect.
  • Unified Framework: The new quadrature representation provides a unified and more intuitive framework for analyzing and designing controllers for a broad class of linear quantum optical systems, including those with squeezing and phase modulation.
  • Systematic Design: The proposed controller parameterization and two-stage optimization approach offer a systematic way to design controllers, ensuring physical realizability and leading to demonstrably improved performance.
  • Performance Gains: The paper demonstrates that by carefully designing controllers using phase modulation and squeezing, substantial improvements in LQG performance can be achieved, surpassing previously reported results.

Key Results

The paper presents a numerical example demonstrating the effectiveness of its approach. By optimizing the controller parameters, including phase modulation angles, the authors achieved a significant reduction in the LQG performance index. Specifically, they report achieving an LQG performance of 3.2515, which is substantially better than the previously reported best coherent LQG control performance of 4.1793 and the baseline performance of 4.8468. This improvement is attributed to the intelligent use of phase modulation and squeezing in the controller design.

Limitations and Future Work

  • Linear Systems: The current framework is limited to linear quantum optical systems. Extending this to nonlinear systems would be a significant next step.
  • Specific Noise Models: While the paper addresses quantum noise, the performance of the controller might be sensitive to the specific noise characteristics and models used.
  • Computational Complexity: The two-stage optimization approach, while effective, can be computationally intensive, especially for systems with a large number of parameters.
  • Future Work: The authors suggest that their controller parameterization and two-stage optimization procedure could be useful for the optimal design of quantum optical networks more broadly. Further exploration of different optimization strategies and applications to more complex quantum systems are implied.

Summary updated

Abstract

This paper studies feedback quantum control of a class of linear quantum optical systems by means of squeezing and phase modulation. A new quadrature representation of such systems is introduced, which includes explicitly phase modulation of light fields involved. Some fundamental relations are derived in terms of the new quadrature representation. The coherent quantum LQG control studied in is re-investigated base on controller parametrization and a two-stage optimization approach. Examples are given to illustrate the effectiveness of the results in the paper.
BibTeX
@misc{zhang2011feedback,
  author = {Hu Zhang and Guofeng Zhang and Daoyi Dong and Bo Huang and Joseph H. W. Lee},
  title = {{Feedback control of linear quantum optical systems}},
  year = {2011},
  doi = {10.1109/ccdc.2011.5968660},
}
Figure