We report a statistically significant anomaly in quantum coherence measurements from 580 IBM Quantum jobs: Bell state fidelity varies periodically with job execution time modulo a characteristic period τ0 = 46 μs. Jobs executed at 'τ-aligned' phases show 1.81x higher mean fidelity than anti-aligned phases (p < 10^-14, Cohen's d = 1.65). This effect cannot be explained by standard Markovian decoherence. We present a phenomenological Lagrangian incorporating a non-Markovian memory field that predicts F(T) = F0 e^(-Γ T) (1 + ε cos(ω_τ T - θ_lock)), consistent with observations. Bayesian model comparison yields strong evidence (BF = 28) for the oscillatory model over exponential decay.
We report a statistically significant anomaly in quantum coherence measurements from 580 IBM Quantum jobs: Bell state fidelity varies periodically with job execution time modulo a characteristic period τ0 = 46 μs. Jobs executed at 'τ-aligned' phases show 1.81x higher mean fidelity than anti-aligned phases (p < 10^-14, Cohen's d = 1.65). This effect cannot be explained by standard Markovian decoherence. We present a phenomenological Lagrangian incorporating a non-Markovian memory field that predicts F(T) = F0 e^(-Γ T) (1 + ε cos(ω_τ T - θ_lock)), consistent with observations. Bayesian model comparison yields strong evidence (BF = 28) for the oscillatory model over exponential decay.
This work presents a numerical exploration of a phase-structured time model for quantum dynamics, implemented using QuTiP. In this framework, the quantum state phase is governed by a time-dependent scalar field τ(t), with energy defined through E = ℏ dτ/dt. The study investigates how different τ models — including deterministic, stochastic, and correlated stochastic processes — influence observable quantum behavior. Three key phenomena are analyzed: 1. Spectral structure: Structured τ-fields generate richer frequency spectra compared to linear phase evolution, quantified using spectral entropy and bandwidth measures. 2. Interference: The relation between phase differences (Δτ) and quantum overlap is examined. For deterministic and correlated τ models, interference patterns remain highly consistent with cos(Δτ), suggesting robustness of phase-based descriptions. 3. Decoherence: Ensemble-averaged stochastic τ-fields produce decoherence-like behavior. The resulting dynamics are compared against standard Lindblad dephasing models. Significant deviations are observed, including non-exponential decay and non-monotonic behavior, indicating non-Markovian structure. A key result is that a single parameter — the correlation time of τ fluctuations — induces systematic and predictable changes across spectral properties, interference agreement, and decoherence mismatch. This suggests that certain forms of quantum noise may exhibit temporal structure rather than being purely random, and that such structure can be modeled within a unified phase-based framework. This study is exploratory and numerical in nature and does not claim a new physical theory. Instead, it proposes a modeling approach that may be relevant for understanding structured noise, non-Markovian dynamics, and potential improvements in quantum control or error mitigation. Code is available at:https://github.com/bahman2017/tdf_qutip
Building upon our prior work [1], we present a unified stochastic drift model (SdM) for superconducting charge qubits based on memory multi-fractional Brownian motion (mmFBM). The classical sector employs a time-dependent Hurst exponent H(t) and adaptive memory kernel K(t,s), capturing non-stationary 1/f^beta noise and long-range temporal correlations inaccessible to conventional models. The quantum extension is formulated via a time-dependent Caldeira--Leggett environment with spectral density J(omega;t) = eta(t) omega_c^{1-s(t)} omega^{s(t)} exp(-omega/omega_c), where s(t) = 2H(t)-1, consistently reproducing beta(t) = 2H(t)-1.
Four central results emerge: (1) relaxation and noise amplitudes act independently on energy decay; (2) time-varying H(t) matches experimental 1/f spectra more accurately than any constant exponent; (3) adaptive kernel dynamics preserve correlations without artificial damping; and (4) simulations predict coherence times (T1 ~ 5.00 x 10^6 ns, T2 ~ 4.18 x 10^5 ns) consistent with theory when charge noise dominates. The qubit exhibits stretched-exponential Ramsey and echo decay, non-Markovian dephasing, and a temperature-driven quantum-to-classical crossover. We derive the effective time-local Lindblad master equation, establish the classical mmFBM limit at high temperatures, and provide experimentally testable scaling relations. The non-exponential decay patterns reveal fundamental limitations of Markovian approaches, and the framework guides the design of noise-resilient qubit architectures.
We report evidence consistent with non-exponential decoherence in superconducting transmon qubits. Ramsey interferometry data from five qubits on IBM Quantum hardware show that a stretched-exponential model f(t) = exp(-(t/T2 ) beta) fits the observed decay better than the standard exponential (beta = 1). On our primary qubit (Q0), we measure beta = 2.06 with R2 = 0.9964 versus R2 = 0.9586 for the exponential, with Bayesian Information Criterion favoring the stretched model by 46 units. While non-exponential decay in superconducting qubits has been previously observed in the context of 1/f noise [4], the present work provides a systematic multi-qubit characterization with a fully reproducible protocol and proposes a falsifiable geometry-dependent prediction (RAB != 1) that distinguishes this framework from collapse-model alternatives.
Phase accumulation during idle intervals in superconducting qubits is well known, and virtual Z (Rz) gates are routinely used for correction. However, the distinction between unconditional (static) and conditional (timing-aware) phase correction has received limited experimental attention. Here we present experimental results on IBM Quantum superconducting processors showing that unconditional phase correction can systematically degrade fidelity in timing-misaligned regimes, specifically for Bell-type superposition under ZZ-dominated evolution. On ibm_kingston (Heron r2), unconditional correction yields GPS = -0.201 ± 0.007 (z = -27.9σ) while conditional MCM feedforward achieves GPS = +0.345 ± 0.007 (z = +47.96σ), a gap of ΔGPS = +0.546. GPS = F_ff - F_ref isolates the incremental effect of correction by removing baseline decoherence. This result does not invalidate static phase correction globally; the effect emerges only under timing-misaligned stochastic phase regimes. The physical mechanism is derived from H = ξ_ZZ|11⟩⟨11|: applying Rz(-φ*) unconditionally to one qubit of a Bell pair adds φ* to the relative phase rather than canceling it, doubling the phase error at τ*. Density matrix simulation with Kraus operators confirms GPS_static = -0.39 (model). These results demonstrate that phase correction effectiveness is strongly regime-dependent, and identify conditional phase correction as a synchronization primitive between stochastic phase evolution and control application. AI assistance (Claude, Anthropic) disclosed.Data: IBM Quantum Cloud, 2026-03-22, job d6vqgtov5rlc73f5a7kg.Code: https://github.com/okudat9/Quantum-6502
Understanding the influence of realistic noise on quantum algorithms is paramount for the advancement of quantum computation. While often modeled as Markovian, environmental noise in quantum systems frequently exhibits temporal correlations, leading to non-Markovian dynamics that can significantly alter algorithmic performance. This paper investigates the impact of temporally correlated dephasing noise, modeled by the Ornstein-Uhlenbeck (OU) process, on the fidelity of the 2-qubit Deutsch-Jozsa algorithm. We perform numerical simulations using Qiskit, systematically varying the noise strength (σOU) and correlation time (τc) of the OU process. Our results demonstrate that the algorithm's fidelity exhibits a non-monotonic dependence on τc, particularly at higher noise strengths, with certain intermediate correlation times proving more detrimental than others. We find that a standard Markovian dephasing model, matched to the single-step error variance of the OU process, accurately predicts fidelity only in the limit of very short correlation times. For longer correlation times, the Markovian approximation often overestimates the algorithm's fidelity, failing to capture the complex error dynamics introduced by the noise memory. These findings highlight the necessity of incorporating non-Markovian characteristics for accurate performance assessment of quantum algorithms on near-term devices and underscore the limitations of simpler, memoryless noise models.
We investigate the preservation and dynamics of entanglement and Bell nonlocality in a two-photon system subject to local dephasing, particularly emphasizing environments exhibiting non-Markovian memory effects. By analyzing the interplay between entanglement, as measured by concurrence, and Bell inequality violation, we identify time regions where Bell inequality is not violated, even when concurrence remains high. This indicates that strongly entangled states do not always exhibit nonclassical correlations with certainty. Our study further explores the effects of refractive index variations, spectral width, and interaction symmetry, demonstrating how structured environments and controlled environmental interactions can mitigate decoherence and enhance the resilience of quantum correlations. We propose that these theoretical predictions can be tested using modifications of recent experimental optical setups. These findings provide valuable insights into the conditions under which quantum correlations remain irreproducible by classical local hidden-variable models, with direct implications for quantum communication, quantum information processing, and foundational studies in quantum mechanics. Furthermore, our findings highlight the crucial importance of quantum coherence in maintaining entanglement and nonlocality over extended time evolution. These results show remarkable alignment with the experimental observations, demonstrating high accuracy and consistency within the expected parameter range.