Robust applicability of continuous dynamical decoupling to decoherence reduction in longitudinal and transverse-noise settings: The role of anisotropy

Abstract: We analytically evaluate the efficiency of continuous dynamical decoupling (CDD) to curb decoherence in generic qubit setups where diverse sources of noise can be present. Previous theoretical approaches to CDD have mainly focused on its potential to cope with longitudinal fluctuations. Here, the basic scenario tackled with CDD is generalized. Apart from dealing with pure dephasing induced by diagonal noise, we consider the impact of transverse fluctuations, usually present in the practical arrangements. In particular, the implications of anisotropic noisy inputs are studied. Additionally, we analyze the role of the fluctuations in the dressing of the qubit by the CDD field of control: since the driving field is usually switched on through linear ramps of its characteristic parameters, the associated dressing of the original states can be described in terms of noisy Landau-Zener transitions. In our approach, based on a sequence of unitary transformations, the noise entering the system is cast into effective stochastic terms whose spectral characteristics are dependent on the driving parameters. This description allows the design of strategies to mitigate the impact of the fluctuations using controlled changes in the effective-noise properties. Significant robustness of CDD against the generalization of the basic scenario can be achieved through an appropriate choice of the parameters of control.

• The paper demonstrates that high drive frequencies shift noise spectral densities, effectively suppressing decoherence in anisotropic environments.
• It develops an analytical framework using unitary transformations to map both longitudinal and transverse noise effects in qudit systems.
• Numerical simulations confirm enhanced coherence times and robust state preparation during adiabatic operations, especially under non-white noise.

Robustness of Continuous Dynamical Decoupling Under Longitudinal and Transverse Noise: The Impact of Anisotropy

Overview

This paper investigates the performance of continuous dynamical decoupling (CDD) as a decoherence suppression method in quantum systems subjected to both longitudinal and transverse noise, with an emphasis on anisotropic noise environments. The analytical framework is developed for generic qudit (including qubit) setups, encompassing both pure dephasing and population transfer effects. The authors generalize canonical noise models to include arbitrary anisotropy in noise components and analyze decoherence both during stationary quantum operations and during the adiabatic preparation of dressed states via noisy Landau-Zener sweeps.

Generalized Hamiltonian and Noise Modeling

The analysis begins with a generalized multi-level system described by the Hamiltonian comprising the qudit energy splitting and a continuous driving field—both affected by stochastic terms:

• Longitudinal noise: Fluctuations along the quantization axis modeled as Ornstein-Uhlenbeck processes.
• Transverse noise: Fluctuations in orthogonal directions, also modeled as Ornstein-Uhlenbeck processes with arbitrary (potentially anisotropic) correlation functions and strengths.

Through a sequence of unitary transformations to the dressed-state basis (incorporating the rotating wave approximation where justified), the original stochastic Hamiltonian is mapped into effective time-dependent terms:

• Longitudinal noise transforms into transverse noise in the dressed basis.
• Transverse noise splits into dephasing contributions and additional effective fluctuating terms, whose spectral characteristics are modified by the control field’s frequency and amplitude.

The effective spectral densities that dictate decoherence in the dressed frame are shifted and combined by deterministic drive terms. Importantly, the spectra of effective noise terms $\chi_a(t)$ and $\chi_b(t)$ (originating from $\eta_x(t)$ and $\eta_y(t)$) are centered around the drive frequency $\omega_d$, resulting in a suppression of decoherence-inducing low-frequency components when $\omega_d$ exceeds the noise bandwidth.

Analytical Treatment of Decoherence Processes

The core decoherence mechanisms—dephasing and population transfer—are analyzed via perturbative and exact treatments in different noise regimes:

Dephasing

• In the presence of broadband noise ($t \gg \tau_c$), the off-diagonal density matrix elements decay exponentially at a rate set by the spectral density of the effective transverse noise at zero frequency, $S(\chi_a; 0)$. This is substantially suppressed for colored noise by increasing the drive frequency $\omega_d$, but not for white noise, where the spectrum is flat.
• In the static-noise limit, explicit expressions for coherence dynamics reveal qualitatively different behaviors for isotropic vs. anisotropic noise. In isotropic cases, only oscillations at the fundamental drive frequency appear. With anisotropy, terms with doubled frequency ($2\omega_d$) emerge; their amplitude scales with the degree of anisotropy.

Population Transfer

• Transition probabilities between dressed states are governed by the spectral densities of both longitudinal and transverse effective noises evaluated at the dressed-state transition energy $\chi_b(t)$0.
• For colored noise, increasing $\chi_b(t)$1 or the drive amplitude $\chi_b(t)$2 (within the RWA regime) further shifts the effective frequencies outside the dominant noise bandwidth, greatly suppressing population leakage. For white noise, transition probabilities remain unaffected by parameter changes.

Anisotropy and Beyond-RWA Effects

The presence of noise anisotropy introduces oscillatory decoherence effects beyond the predictions of standard RWA treatments. The analysis includes explicit calculations that retain oscillating components at $\chi_b(t)$3 and demonstrates that their impact on system evolution is secondary to the leading deterministic (coherent) corrections, provided transverse noise amplitudes remain small compared to drive frequencies.

Adiabatic Preparation and Noisy Landau-Zener Transitions

The study rigorously incorporates the effect of noise during the adiabatic ramping of the control field—that is, while preparing dressed states via Landau-Zener sweeps with both amplitude and detuning ramps in the presence of noise.

Key findings:

• Fluctuations introduce additional stochastic phases and possible transitions, but population transfer rates are again suppressed if the effective noise spectral densities at instantaneous energy gaps remain small.
• For static or colored noise with narrow bandwidths (relative to drive frequency), negligible non-adiabatic transitions occur, ensuring robust preparation.
• The only exception is the white-noise limit, wherein no parameter regime allows for dynamical suppression due to the absence of a spectral gap.

Numerical Results

Comprehensive numerical simulations corroborate the analytic predictions, demonstrating:

• Exponential dephasing in the regime $\chi_b(t)$4 with rates precisely set by $\chi_b(t)$5 evaluated for various parameter sets.
• Dramatic enhancement of coherence times with increasing drive frequency and/or decreasing noise bandwidth.
• For realistic parameters relevant to ultracold atomic clocks and solid-state qubit implementations, the calculated coherence extension agrees with recent experiments.

Practical and Theoretical Implications

The formalism provided enables the design of CDD protocols optimized for arbitrary combinations of longitudinal and anisotropic transverse noise. It establishes:

• High drive frequencies (subject to being near-resonant with the system) and sufficient drive strength effectively suppress all decoherence provided noise is narrow-band.
• The presence of anisotropy, while introducing higher harmonics in dynamical quantities, does not constitute a principal limitation in the parameter regime of interest, as these components are subdominant.
• During dressed-state preparation, standard adiabatic protocols remain effective unless the ambient noise is genuinely white.

On the theoretical front, the effective noise mapping via unitary transformation elucidates the general conditions under which CDD techniques are robust, even for qudit systems and arbitrary noise directions. The mathematical framework accommodates further extensions, such as inclusion of amplitude fluctuations and non-Gaussian noise.

Prospective Advances

The results pave the way for several avenues:

• Tailored CDD parameter optimization considering full environmental spectral densities, including cross-correlations and higher-order noise statistics.
• Integration with machine learning-based control-field design protocols, especially in multi-qubit (qudit) architectures.
• Direct experimental validation in platforms evidencing strong transverse noise anisotropy, such as hole spin qubits, NV centers, and engineered cold atom arrays.

Conclusion

This work provides an exhaustive analytical framework for understanding and optimizing the performance of continuous dynamical decoupling under generalized noise models, including arbitrary anisotropy and fluctuations during state preparation. The findings establish the method's robustness for both dephasing and population transfer suppression, provided non-white noise spectra and operation within the near-resonant, RWA-valid regime. These insights have direct applicability to the engineering of coherent quantum technologies across atomic, solid-state, and hybrid platforms.

Reference: "Robust applicability of continuous dynamical decoupling to decoherence reduction in longitudinal and transverse-noise settings: The role of anisotropy" (2606.08114)