A fidelity measure for quantum channels

Maxim RaginskyPhysics 2001 · DOI · arXiv

Introduction

In quantum information theory, the ability to quantify the similarity between two quantum states is fundamental. The Jozsa fidelity, $F(\rho, \sigma)$, serves as the standard metric for this purpose. However, quantum information processing is defined not just by static states, but by the transformations they undergo—mathematically represented as quantum channels (completely positive trace-preserving maps).

While the distinguishability of states is well-understood, establishing a physically intuitive and mathematically robust measure for distinguishing two different quantum processes has historically been more complex. This paper, authored by Maxim Raginsky, addresses this gap by introducing a formal fidelity measure for quantum channels. By leveraging the duality between channels and bipartite states, the author provides a framework that extends the properties of state fidelity to the realm of quantum operations, proving a channel-based version of Uhlmann’s theorem and exploring its implications for error correction.

Theoretical Framework: The State-Channel Duality

The core methodology of the paper relies on the Choi-Jamiołkowski isomorphism, a powerful mathematical bridge that creates a one-to-one correspondence between quantum channels and bipartite density operators.

To define the fidelity between two channels $S$ and $T$, the author follows a three-step process:

  1. Mapping Channels to States: A channel $T$ acting on a system $A$ is associated with a specific bipartite state $\rho_T$ on a composite system $A \otimes A$. This is achieved by applying the channel to one half of a maximally entangled state.
  2. Defining Channel Fidelity: The fidelity between two channels $S$ and $T$ is then defined as the state fidelity between their corresponding representative states: $\mathcal{F}(S, T) := F(\rho_S, \rho_T)$.
  3. Stinespring Dilation: To explore the geometric properties of this measure, the paper employs Stinespring dilations. This technique represents a noisy channel as a unitary evolution on a larger, extended Hilbert space (the environment). This allows for the derivation of an analogue to Uhlmann’s Theorem, which posits that channel fidelity is equivalent to the maximum overlap between the "purified" isometries of the channels.

Key Insights and Mathematical Properties

The paper establishes that the properties of state fidelity—such as symmetry, concavity, and multiplicativity—translate directly to channels through this duality. A significant conceptual insight involves the roles of preprocessing and postprocessing in channel distinguishability:

  • Postprocessing: Applying a transformation to the output of channels can only make them harder to distinguish, a property known as monotonicity.
  • Preprocessing: Conversely, preparing a specific input state (potentially entangled with an auxiliary system) can enhance the ability to distinguish between two channels.

Formal Properties

The author validates seven core properties (CF1-CF7) for the proposed measure, including:

  • Identity: $\mathcal{F}(S, T) = 1$ if and only if $S = T$.
  • Multiplicativity: The measure is multiplicative under tensor products: $\mathcal{F}(S_1 \otimes S_2, T_1 \otimes T_2) = \mathcal{F}(S_1, T_1)\mathcal{F}(S_2, T_2)$.
  • Geometric Interpretation: Using "double-ket" notation $|A\rangle\rangle$ to transform operators into vectors, the paper demonstrates that channel fidelity provides a geometric view of how closely two transformations align in the operator space.

Relationship to the Diamond Norm

A critical component of the research is comparing channel fidelity to the norm of complete boundedness (cb-norm), also known as the diamond norm ($\|S - T\|_{cb}$). The diamond norm is often considered the "worst-case" measure of distinguishability. Raginsky provides a specific inequality relating the two:

$$2 - 2\sqrt{\mathcal{F}(S, T)} \leq \|S - T\|_{cb}$$

This result indicates that while the diamond norm is a stricter criterion, channel fidelity offers a useful "average-case" or state-based perspective that is often more mathematically tractable in practical applications.

Applications in Quantum Error Correction

The utility of this measure is demonstrated through its application to Quantum Error-Correcting Codes (QECC). The paper proves that a subspace serves as a perfect quantum error-correcting code for a specific channel if the fidelity between the identity channel and the "recovered" channel (the composition of the noise and the correction procedure) is exactly 1. This provides a rigorous benchmark for evaluating the performance of error-correction protocols.

Limitations and Conclusion

While the proposed fidelity measure is robust, the author notes certain limitations. In high-security cryptographic scenarios, the channel fidelity might be considered too "lenient" compared to the diamond norm. Additionally, while the paper proves that optimal preprocessing can improve distinguishability, finding that optimal input for arbitrary, high-dimensional channels remains a significant computational challenge.

In summary, this work provides a definitive "yardstick" for quantum processes. By unifying state and channel fidelity under a single framework, it simplifies the mathematical toolkit available to quantum information scientists and provides a clear path for analyzing the stability and accuracy of quantum gates and communication lines.

Summary updated

BibTeX
@inproceedings{raginsky2001fidelity,
  author = {Maxim Raginsky},
  title = {{A fidelity measure for quantum channels}},
  booktitle = {Physics},
  year = {2001},
  doi = {10.1016/s0375-9601(01)00640-5},
}
Figure