| entanglement-assisted classical capacity | |
|---|---|
| Name | Entanglement-assisted classical capacity |
| Field | Quantum information theory |
| Introduced | 1990s |
| Related | Holevo bound, Schumacher–Westmoreland theorem, quantum channel capacity |
entanglement-assisted classical capacity
Entanglement-assisted classical capacity is the maximum rate at which classical information can be reliably transmitted over a quantum communication channel when the sender and receiver share unlimited prior entanglement. It refines notions from classical information theory and quantum information theory by quantifying how pre-shared quantum entanglement acts as a resource to boost classical communication, with operational significance for quantum networks and protocols such as superdense coding and quantum error correction.
Entanglement-assisted classical capacity (often abbreviated EA capacity) formalizes the achievable classical data rate per channel use when unlimited prior entanglement between sender (Alice) and receiver (Bob) is available. The concept builds on the Shannon notion of capacity and the Holevo bound for quantum ensembles, and it is central to resource theories in quantum communication. The EA capacity is typically denoted C_E or C_{EA} for a given quantum channel (completely positive trace-preserving map) and is expressed in bits per channel use using the von Neumann entropy and quantum mutual information. Key contributors to the theory include Alexander Holevo, Benjamin Schumacher, and Michael Westmoreland.
The formalism treats a quantum channel 𝒩 : 𝔅(ℋ_A) → 𝔅(ℋ_B) as the primitive. Unlimited shared entanglement is modeled by a pre-shared pure state such as many copies of a maximally entangled state (e.g., Bell states or EPR pairs) between ancilla systems at Alice and Bob. Encoding is performed by local operations conditioned on classical messages; decoding uses joint measurements assisted by the entangled ancilla. The EA capacity is derived using the quantum generalization of mutual information, S(ρ) for von Neumann entropy, and the quantum mutual information I(A;B)_σ = S(σ_A)+S(σ_B)-S(σ_{AB}). The optimization often ranges over input density operators and purification choices, invoking properties of completely positive trace-preserving maps and the Stinespring dilation theorem.
The Holevo–Schumacher–Westmoreland theorem (HSW theorem) characterizes the unassisted classical capacity of quantum channels via the Holevo information under product-state encodings and regularization. For entanglement assistance, a landmark result gives a single-letter formula: the entanglement-assisted classical capacity equals the maximum quantum mutual information across the channel, C_{EA}(𝒩) = max_{ρ} I(A;B)_{(id⊗𝒩)(Φ_{A A'})}, where Φ is a purification of ρ. This formula parallels the quantum mutual information used in quantum teleportation and is additive, avoiding the infinite regularization needed in some unassisted capacities. The proof techniques exploit typical subspace methods, packing bounds, and entanglement-assisted coding theorems developed by Holevo, Schumacher, Westmoreland, and later formalized in works by Bennett and collaborators.
For specific channels the EA capacity can be computed analytically. For an ideal noiseless qubit channel (identity channel) with input dimension d, C_{EA} = 2 log_2 d bits per channel use, reflecting the superdense coding factor of two when maximally entangled pairs are available. For the depolarizing channel and erasure channel, closed-form expressions exist: the erasure channel with erasure probability p has C_{EA} = (1-p) log_2 d + H((1-p)) adjustments via conditional entropies, while the depolarizing channel requires optimization over input ensembles and often yields symmetric extremizers. Bosonic Gaussian channels such as the lossy bosonic channel and thermal noise channel admit EA capacities expressed using continuous-variable entropy formulas, relating to works by Giovannetti, Lloyd, and Shapiro.
Operational implementations of entanglement-assisted communication include protocols combining pre-shared entanglement with classical encodings and collective decoding; paradigmatic examples are superdense coding and variants using entanglement distillation and entanglement swapping to generate the required resource. Trade-offs arise when entanglement is finite: achievable rate regions relate classical communication, quantum communication, and entanglement consumption, formalized in multi-resource quantum Shannon theory as trade-off curves (e.g., classical-quantum-entanglement capacities). Practical strategies often interleave quantum error correction and entanglement purification to mitigate channel noise and finite-storage constraints.
Unlike many unassisted capacities that require regularization due to non-additivity (examples include the quantum capacity and some forms of private capacity), the entanglement-assisted classical capacity is additive and single-letter, simplifying both conceptual understanding and computation. This additivity follows from subadditivity properties of the quantum mutual information and avoids pathological superadditivity examples encountered in capacities like the Holevo capacity. The EA result underscores how shared entanglement linearizes certain channel behaviors and connects to broader additivity conjectures historically studied by researchers at institutions such as IBM Research, MIT, and Caltech.
Experimental realizations of entanglement-assisted communication demand high-fidelity entanglement generation and distribution, low-loss channels, and coherent joint measurements. Platforms include trapped ions (e.g., experiments at University of Innsbruck), superconducting circuits (e.g., Google Quantum AI, IBM Quantum), photonic systems in free-space and fiber networks (demonstrations by NIST and university groups), and continuous-variable implementations using optical parametric amplifiers by groups led by Sergio L. Braunstein and Stefano Pirandola. Challenges include scaling entanglement distribution via quantum repeaters, integrating quantum key distribution modules, and mitigating decoherence. Benchmarks compare EA capacity estimates to achievable rates in laboratory settings, guiding development of quantum network standards and protocols for future quantum internet architectures.
Category:Quantum information theory Category:Quantum communication