Bell-Inequality Violation for Continuous, Non-Projective Measurements

Technical Summary: Bell-Inequality Violation for Continuous, Non-Projective Measurements

Problem Statement
Standard experimental demonstrations of Bell inequality violations rely on sharp, projective, dichotomic measurements performed on well-isolated two-level systems. However, many emerging solid-state quantum platforms—such as superconducting circuits under continuous dispersive readout, spin ensembles, and driven-dissipative quantum materials—do not support such idealized measurements. Instead, these systems yield continuous voltage or field traces characterized by weak, non-demolition readout. In these regimes, the measurement does not collapse the quantum state into a discrete eigenbasis but produces stochastic trajectories encoding partial information about the underlying quantum phase structure. This creates a fundamental gap: standard Bell tests are not directly applicable to continuous time-series data, raising the question of how quantum nonlocality can be certified without assuming a specific collapse model or phase distribution.

Methodology
The authors develop a general theoretical framework to extract Bell–CHSH inequality violations directly from continuous, non-projective measurement records. The methodology proceeds through the following steps:

1. Phase-Resolved Representation: The framework treats the continuous measurement record as a source of a phase variable $\phi_k(t)$ for each subsystem $k$. Rather than assuming a hidden classical variable, the phase is treated as a label for a continuous family of measurement outcomes. The joint statistics are described by a phase-resolved amplitude kernel $a(\phi_1, \phi_2)$.
2. Dimensionality Reduction: The authors demonstrate that despite the apparent infinite dimensionality of continuous records, the relevant measurement statistics for Bell correlations collapse onto a universal two-dimensional subspace. This subspace is spanned by the constant mode and the second harmonic of the phase ($e^{i2\phi}$), dictated by the $\pi$-periodicity and second-harmonic sensitivity of CHSH observables.
3. Construction of Dichotomic Observables: Within this reduced subspace, the authors define operational dichotomic observables $\hat{A}(A)$ and $\hat{B}(B)$ parameterized by external angles. These are constructed via orthogonal projection of the phase-resolved states, accounting for the local phase distribution $P(\phi)$.
4. Derivation of the Correlator: The Bell correlator $E(A, B)$ is derived as a function of two distinct physical resources:
   • Local Phase Coherence ($\kappa(P)$): A factor quantifying the intrinsic phase uncertainty of individual subsystems, derived from the second angular moment of the local phase distribution $P(\phi)$.
   • Nonlocal Phase Locking ($\gamma$): A complex coherence parameter derived from the second Fourier component of the phase-difference distribution (the phase-locking kernel $E(\Delta\phi)$).
5. Classical Null Hypothesis: The paper establishes a rigorous classical bound by defining a null hypothesis based on stationary local stochastic models. It proves that any classical model, where measurement outcomes are generated locally and deterministically from shared hidden variables, satisfies the CHSH bound $|S| \leq 2$ when processed through the same estimator.

Key Contributions

• Operational Framework for Continuous Data: The paper provides a principled route to map continuous, weak measurement records onto the CHSH formalism without requiring discrete projective outcomes.
• Identification of Necessary Resources: The work identifies that Bell–CHSH violation in continuous systems requires the joint presence of two resources: finite intrinsic phase uncertainty (local coherence, $\kappa$) and coherent phase locking between subsystems (nonlocal coherence, $\gamma$). If either resource vanishes (e.g., predetermined phase or no phase locking), the correlations revert to classical bounds.
• Closed-Form Estimator: For Gaussian phase-locking models, the authors derive a closed-form expression for the correlator: $E(A, B) = \kappa(P) \gamma \cos(2(A-B))$. This leads to a specific violation condition: $|\kappa(P) \gamma| > 1/\sqrt{2}$.
• Validation via Simulation: The resulting estimator is benchmarked against conventional projective-measurement CHSH tests implemented via Qiskit quantum-circuit simulations. The study finds quantitative agreement in the Bell-violating regime without parameter fitting, confirming that the reduced-phase estimator reliably tracks entanglement.

Results

• Classical vs. Quantum Behavior: The analysis confirms that purely diagonal phase statistics (classical hidden-variable models) cannot violate the CHSH inequality. In contrast, quantum phase-locked systems recover the nonlinear angular dependence characteristic of entanglement.
• Quantitative Agreement: In the regime where entanglement is present, the continuous-measurement estimator tracks the Qiskit benchmark with high accuracy, crossing the classical bound $S=2$ at the same correlation scale (phase-lock variance $\sigma_L$).
• Conservative Bounds: Below the classical threshold, the estimator exhibits a more rapid decay than the projective benchmark, indicating it provides a conservative bound on nonclassical correlations suitable for entanglement detection.

Significance and Claims
The paper claims to provide a practical route for demonstrating Bell nonlocality in platforms where measurements are inherently continuous and weak, such as superconducting circuits and polariton condensates. The significance of the work lies in:

• Extending Bell-type analysis to continuously monitored many-body quantum platforms where traditional projective tests are impractical.
• Clarifying that nonlocality in these systems is not a monolithic property but arises from the interplay of local phase uncertainty and nonlocal phase locking.
• Establishing a rigorous classical benchmark for a well-defined class of continuous-record estimators. The authors explicitly state that this framework does not claim device-independent nonlocality in the strict, loophole-free sense. Instead, it offers a transparent, assumption-based method to identify correlations that cannot be reproduced by local classical stochastic processes processed through the same measurement pipeline.

The work concludes that by explicitly making the physical assumptions underlying the measurement process and the classical null hypothesis, this framework enables the assessment of nonclassical correlations in regimes previously inaccessible to standard Bell tests.