arXiv:2610.20725
Quantum prepare-and-measure scenarios: concepts and methods
A. de Oliveira Junior, Santiago Zamora, Jonatan Bohr Brask, Rafael Chaves, Armin Tavakoli
Abstract
Prepare-and-measure scenarios are among the simplest and most widely used settings for quantum communication. A sender prepares a physical system in one of several possible states, which a receiver then measures. Despite this simple structure, these scenarios offer a rich framework for exploring the differences between classical and quantum communication, revealing quantum advantages, and developing protocols for quantum cryptography and device certification. In this tutorial, we give a ground-up introduction to prepare-and-measure scenarios and the methods used to study them. We introduce different frameworks for constraining and characterising communication, with an emphasis on semi-device-independent approaches to quantum information. We also cover modern convex-optimisation techniques for characterising the correlations achievable in these scenarios and illustrate how these tools can be applied in practice. Our aim is to give newcomers the conceptual and technical foundations needed to approach the current literature and apply prepare-and-measure methods in their own research. We conclude with a discussion of open problems and directions for future work.
AI Summary
This tutorial asks how restrictions on communication make prepare-and-measure statistics $p(b|x,y)$ distinguish classical from quantum resources, and how those distinctions support device certification and semi-device-independent cryptography. It develops the framework from dimension-bounded classical and quantum messages with shared randomness, then covers entanglement assistance, networks, and alternative constraints involving energy, fidelity, overlaps, information content, and small high-dimensional leakage. It synthesises established results rather than presenting a new central theorem. Worked examples explain classical correlation polytopes, quantum random access codes, dimension witnesses, self-testing, and certification of nonprojective and entangled measurements. Key benchmarks include the Frenkel–Weiner theorem: with one receiver measurement and shared randomness, quantum communication has no advantage over classical communication of equal dimension. For the two-bit random access code, a classical bit achieves at most $3/4$, whereas a qubit achieves $(1+1/\sqrt{2})/2\approx0.8536$. More generally, the two-symbol, $d$-dimensional code has quantum optimum $(1+1/\sqrt d)/2$, compared with classical optimum $(1+1/d)/2$. Derivations use deterministic-strategy enumeration, eigenvalue optimisation, geometric arguments, and equality conditions for certification. The computational treatment covers linear-programming membership tests, alternating state–measurement optimisation, moment-matrix semidefinite relaxations, and symmetry reduction. Explicit limitations include local optima in seesaw searches, heuristic sampling in some dimension-constrained relaxations, rapidly growing computational costs, and poor noise robustness of known nonprojective-measurement self-tests. The closing research agenda includes rigorous scalable correlation bounds, security with memory or shared entanglement, loss tolerance, and comparisons between communication assumptions.
Error found: The general self-testing witness preceding Eq. (miguelselftest) incorrectly uses ket-difference norms instead of projector-difference trace norms, making its coefficients phase-dependent and invalidating its stated saturation property.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 78 pages, 15 figures. Comments welcome