Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2606.00358 · PRX Quantum

Software-based compensation of AC-line-induced control errors in qubits and qudits

Gaurav A. Tathed, Nicholas C. F. Zutt, Collin J. C. Epstein, Crystal Senko

quant-phSubmitted May 29, 2026• Announced Jun 2, 2026• 1 citation
v1 · May 29, 2026v2 · Jun 23, 2026
arXivHTMLPDF

Abstract

AC mains power line-synchronous disturbances are a common source of coherent, time-dependent error in precision quantum-control experiments. We show that when these disturbances are reproducible with respect to the mains phase, their effect can be measured in a line-triggered frame and compensated through software updates to control sequences. In our system, the disturbances manifest as magnetic-field-induced shifts in the energy level structure of a trapped $^{137}\text{Ba}^+$ ion, resulting in time-dependent detunings between the ion transitions and a local oscillator, as well as additional phases accumulated on superpositions of energy levels. We demonstrate a compensation protocol that corrects for the instantaneous oscillator detuning during control pulses, and for the phase accumulated by the energy levels between pulses. The calibrated AC line contribution to the detuning is reduced by $21(9)\times$, while the fitted AC phase amplitude is reduced below the measurement uncertainty. We then study gate performance on a magnetic-field-sensitive qubit and find that uncompensated mains-synchronous errors produce time-dependent fluctuations that make the usual randomized-benchmarking decay model unreliable. With compensation enabled, these fluctuations are suppressed sufficiently to recover a standard benchmarking decay and extract an average gate fidelity of $99.93(1)\%$. Finally, we extend the framework to multilevel qudit control and apply it to a single-qudit Bernstein-Vazirani algorithm, where AC compensation increases the success probability on a 16-level qudit from $10(7)\%$ to $70(9)\%$. These results show that reproducible line-synchronous noise can be treated as a calibrated control-frame error and corrected without additional hardware.

Comments
18 pages, 4 figures

Related papers

Calibrating magnetic flux control in superconducting circuits by compensating distortions on timescales from nanoseconds up to tens of microseconds

Christoph Hellings, Nathan Lacroix, Ants Remm +8 more

Fast tuning of the transition frequency of superconducting qubits using magnetic flux is essential, for example, for realizing high-fidelity two-qubit gates with low leakage or for reducing errors in dispersive qubit readout. To apply accurately shaped flux pulses, signal distortions induced by the flux control lines need to be carefully compensated for. This requires their characterization at the reference plane of the qubit. However, many existing approaches are limited in time resolution or in pulse duration. Here, we overcome these limitations and demonstrate accurate flux control with subpermille residual frequency errors on timescales ranging from nanoseconds to tens of microseconds. We achieve this by combining two complementary methods to characterize and compensate for pulse distortions. We have deployed this approach successfully in a quantum error correction experiment calibrating 24 flux-activated two-qubit gates. Reliable calibration methods, as the ones presented here, are essential in experiments scaling up superconducting quantum processors.

arXiv:2503.04610•Physical Review Research•Nov 7, 2025•21 citations

Quantum_Control_System_with_Active_Thermal_Compensation

Walter Ayala

This document develops a theoretical, technical, and experimental framework for activethermal compensation of quantum-control pulses in superconducting qubits. The centralproposal is to measure a temperature-related control variable and use that information toadapt the microwave control amplitude so that the realized gate rotation is less sensitive toslow or reproducible thermal variations.Version 2.0 makes an important distinction between established physical relations andphenomenological assumptions. In particular, no universal linear dependence of the Rabifrequency on temperature is assumed. Instead, the temperature-to-control response is rep-resented by experimentally identifiable transfer functions and local calibration coefficients.Possible microscopic contributions from quasiparticles, thermal photons, two-level systems,dielectric loss, and other environmental channels are treated as candidate mechanisms ratherthan as a universal additive law.The control problem is formulated using a driven two-level Hamiltonian with both am-plitude and detuning errors. A thermal sensitivity coefficient is introduced through theexperimentally measurable derivative of the effective Rabi frequency with respect to tem-perature. The compensation law is derived from this measured response. Finite sensor andactuator bandwidth are incorporated through a causal filter and an explicit latency term.The experimental protocol combines Rabi calibration, thermal characterization, random-ized benchmarking, interleaved randomized benchmarking, and independent measurementsof T1 and T∗2. The primary experimental question is whether active thermal compensationproduces a statistically and experimentally resolvable reduction of gate error under controlledthermal perturbations.The proposal does not assume a predetermined improvement in fidelity, and it does notassume that amplitude compensation automatically suppresses all thermal decoherence. Allnumerical coefficients introduced in this document are design parameters or initial experi-mental ranges unless explicitly identified as measured quantities. No experimental result isclaimed.

•Sep 20, 2026•0 citations

Compensation of power line-induced magnetic interference in trapped-ion system

Hu Han, Yi Xie, Man-chao Zhang +7 more

Abstract The low-frequency magnetic field noise resulting from the lab[1]oratory power supply is one of the main decoherence sources of trapped-ion qubits. Synchronizing the experimental sequence to the phase of the power line is widely used to mitigate this problem, but it will greatly reduce the ex[1]perimental efficiency. In this paper, we experimentally demonstrate a simple active compensation method to reduce the observed 50 Hz and 150 Hz strong magnetic noise in an ion trap induced by the power line. In our method, a single 40Ca+ ion is used as the noise probe and an reverse compensation sig[1]nal is generated by a programmable arbitrary waveform generator (AWG).After compensation, an 86% reduction of the periodic magnetic field fluc[1]tuation and over 35-fold extension of the coherence time from 70 µs to 2500 µs were observed. This method can also be applied to compensate other spectral components of the magnetic field noise related to the power line, and it is also useful for other atomic systems such as neutral atoms.

Applied Physics B•Oct 4, 2023•5 citations

Compensating sequences for robust quantum control of trapped-ion qubits

J. True Merrill

Universal quantum computation requires precision control of the dynamics of qubits. Frequently accurate quantum control is impeded by systematic drifts and other errors. Compensating composite pulse sequences are a resource efficient technique for quantum error reduction. This work describes compensating sequences for ion-trap quantum computers. We introduce a Lie-algebraic framework which unifies all known fully-compensating sequences and admits a novel geometric interpretation where sequences are treated as vector paths on a dynamical Lie algebra. Using these techniques, we construct new narrowband sequences with improved error correction and reduced time costs. We use these sequences to achieve laser addressing of single trapped 40Ca+ ions, even if neighboring ions experience significant field intensity. We also use broadband sequences to achieve robust control of 171Yb+ ions even with inhomogeneous microwave fields. Further, we generalize compensating sequences to correct certain multi-qubit interactions. We show that multi-qubit gates may be corrected to arbitrary accuracy if there exists either two non-commuting controls with correlated errors or one error-free control. A practical ion-trap quantum computer must be extendible to many trapped ions. One solution is to employ microfabricated surface-electrode traps, which are well-suited for scalable designs and integrated systems. We describe two novel surface-electrode traps, one with on-chip microwave waveguides for hyperfine 171Yb+ qubit manipulations, and a second trap with an integrated high numerical aperture spherical micromirror for enhanced fluorescence collection.

•May 17, 2013•2 citations

Composite pulses with errant phases

Boyan T. Torosov, Nikolay V. Vitanov

Composite pulses---sequences of pulses with well-defined relative phases---are an efficient, robust, and flexible technique for coherent control of quantum systems. Composite sequences can compensate for a variety of experimental errors in the driving field (e.g., in the pulse amplitude, duration, detuning, chirp, etc.) or in the quantum system and its environment (e.g., inhomogeneous broadening, stray electric or magnetic fields, unwanted couplings, etc.). The control parameters are the relative phases between the constituent pulses in the composite sequence, an accurate control over which is required in all composite sequences reported hitherto. In this paper, we introduce two types of composite pulse sequences which, in addition to error compensation in the basic experimental parameters, compensate for systematic errors in the composite phases. In the first type of such composite sequences, which compensate for pulse area errors, relative phase errors of over 10% can be tolerated with reasonably short sequences while maintaining the fidelity above the 99.99% quantum computing benchmark. In the second type of composite sequences, which compensate for simultaneous pulse area and detuning errors, relative phase errors of over 5% can be compensated.

arXiv:1904.13168•Physical Review A•Aug 12, 2019•23 citations

In situ characterization of qubit-drive phase distortions

Mario F. Gely, J.M.A. Litarowicz, A. D. Leu +1 more

Reducing errors in quantum gates is critical to the development of quantum computers. To do so, any distortions in the control signals should be identified; however, conventional tools are not always applicable when part of the system is under high vacuum, cryogenic, or microscopic. Here, we demonstrate a method to detect and compensate for amplitude-dependent phase changes, using the qubit itself as a probe. The technique is implemented using a microwave-driven trapped-ion qubit, where correcting phase distortions leads to a threefold improvement in the error of single-qubit gates implemented with pulses of different amplitudes, to attain state-of-the-art performance benchmarked at 1.6 ( 4 ) × 10 − 6 error per Clifford gate. Published by the American Physical Society 2024

arXiv:2309.14703•Physical Review Applied•Aug 1, 2024•5 citations
AC mains power line-synchronous disturbances are a common source of coherent, time-dependent error in precision quantum-control experiments. We show that when these disturbances are reproducible with respect to the mains phase, their effect can be measured in a line-triggered frame and compensated through software updates to control sequences. In our system, the disturbances manifest as magnetic-field-induced shifts in the energy level structure of a trapped 137Ba+^{137}\text{Ba}^+137Ba+ ion, resulting in time-dependent detunings between the ion transitions and a local oscillator, as well as additional phases accumulated on superpositions of energy levels. We demonstrate a compensation protocol that corrects for the instantaneous oscillator detuning during control pulses, and for the phase accumulated by the energy levels between pulses. The calibrated AC line contribution to the detuning is reduced by 21(9)×21(9)\times21(9)×, while the fitted AC phase amplitude is reduced below the measurement uncertainty. We then study gate performance on a magnetic-field-sensitive qubit and find that uncompensated mains-synchronous errors produce time-dependent fluctuations that make the usual randomized-benchmarking decay model unreliable. With compensation enabled, these fluctuations are suppressed sufficiently to recover a standard benchmarking decay and extract an average gate fidelity of 99.93(1)%99.93(1)\%99.93(1)%. Finally, we extend the framework to multilevel qudit control and apply it to a single-qudit Bernstein-Vazirani algorithm, where AC compensation increases the success probability on a 16-level qudit from 10(7)%10(7)\%10(7)% to 70(9)%70(9)\%70(9)%. These results show that reproducible line-synchronous noise can be treated as a calibrated control-frame error and corrected without additional hardware.
Fast tuning of the transition frequency of superconducting qubits using magnetic flux is essential, for example, for realizing high-fidelity two-qubit gates with low leakage or for reducing errors in dispersive qubit readout. To apply accurately shaped flux pulses, signal distortions induced by the flux control lines need to be carefully compensated for. This requires their characterization at the reference plane of the qubit. However, many existing approaches are limited in time resolution or in pulse duration. Here, we overcome these limitations and demonstrate accurate flux control with subpermille residual frequency errors on timescales ranging from nanoseconds to tens of microseconds. We achieve this by combining two complementary methods to characterize and compensate for pulse distortions. We have deployed this approach successfully in a quantum error correction experiment calibrating 24 flux-activated two-qubit gates. Reliable calibration methods, as the ones presented here, are essential in experiments scaling up superconducting quantum processors.
This document develops a theoretical, technical, and experimental framework for activethermal compensation of quantum-control pulses in superconducting qubits. The centralproposal is to measure a temperature-related control variable and use that information toadapt the microwave control amplitude so that the realized gate rotation is less sensitive toslow or reproducible thermal variations.Version 2.0 makes an important distinction between established physical relations andphenomenological assumptions. In particular, no universal linear dependence of the Rabifrequency on temperature is assumed. Instead, the temperature-to-control response is rep-resented by experimentally identifiable transfer functions and local calibration coefficients.Possible microscopic contributions from quasiparticles, thermal photons, two-level systems,dielectric loss, and other environmental channels are treated as candidate mechanisms ratherthan as a universal additive law.The control problem is formulated using a driven two-level Hamiltonian with both am-plitude and detuning errors. A thermal sensitivity coefficient is introduced through theexperimentally measurable derivative of the effective Rabi frequency with respect to tem-perature. The compensation law is derived from this measured response. Finite sensor andactuator bandwidth are incorporated through a causal filter and an explicit latency term.The experimental protocol combines Rabi calibration, thermal characterization, random-ized benchmarking, interleaved randomized benchmarking, and independent measurementsof T1 and T∗2. The primary experimental question is whether active thermal compensationproduces a statistically and experimentally resolvable reduction of gate error under controlledthermal perturbations.The proposal does not assume a predetermined improvement in fidelity, and it does notassume that amplitude compensation automatically suppresses all thermal decoherence. Allnumerical coefficients introduced in this document are design parameters or initial experi-mental ranges unless explicitly identified as measured quantities. No experimental result isclaimed.
Abstract The low-frequency magnetic field noise resulting from the lab[1]oratory power supply is one of the main decoherence sources of trapped-ion qubits. Synchronizing the experimental sequence to the phase of the power line is widely used to mitigate this problem, but it will greatly reduce the ex[1]perimental efficiency. In this paper, we experimentally demonstrate a simple active compensation method to reduce the observed 50 Hz and 150 Hz strong magnetic noise in an ion trap induced by the power line. In our method, a single 40Ca+ ion is used as the noise probe and an reverse compensation sig[1]nal is generated by a programmable arbitrary waveform generator (AWG).After compensation, an 86% reduction of the periodic magnetic field fluc[1]tuation and over 35-fold extension of the coherence time from 70 µs to 2500 µs were observed. This method can also be applied to compensate other spectral components of the magnetic field noise related to the power line, and it is also useful for other atomic systems such as neutral atoms.
Universal quantum computation requires precision control of the dynamics of qubits. Frequently accurate quantum control is impeded by systematic drifts and other errors. Compensating composite pulse sequences are a resource efficient technique for quantum error reduction. This work describes compensating sequences for ion-trap quantum computers. We introduce a Lie-algebraic framework which unifies all known fully-compensating sequences and admits a novel geometric interpretation where sequences are treated as vector paths on a dynamical Lie algebra. Using these techniques, we construct new narrowband sequences with improved error correction and reduced time costs. We use these sequences to achieve laser addressing of single trapped 40Ca+ ions, even if neighboring ions experience significant field intensity. We also use broadband sequences to achieve robust control of 171Yb+ ions even with inhomogeneous microwave fields. Further, we generalize compensating sequences to correct certain multi-qubit interactions. We show that multi-qubit gates may be corrected to arbitrary accuracy if there exists either two non-commuting controls with correlated errors or one error-free control. A practical ion-trap quantum computer must be extendible to many trapped ions. One solution is to employ microfabricated surface-electrode traps, which are well-suited for scalable designs and integrated systems. We describe two novel surface-electrode traps, one with on-chip microwave waveguides for hyperfine 171Yb+ qubit manipulations, and a second trap with an integrated high numerical aperture spherical micromirror for enhanced fluorescence collection.
Composite pulses---sequences of pulses with well-defined relative phases---are an efficient, robust, and flexible technique for coherent control of quantum systems. Composite sequences can compensate for a variety of experimental errors in the driving field (e.g., in the pulse amplitude, duration, detuning, chirp, etc.) or in the quantum system and its environment (e.g., inhomogeneous broadening, stray electric or magnetic fields, unwanted couplings, etc.). The control parameters are the relative phases between the constituent pulses in the composite sequence, an accurate control over which is required in all composite sequences reported hitherto. In this paper, we introduce two types of composite pulse sequences which, in addition to error compensation in the basic experimental parameters, compensate for systematic errors in the composite phases. In the first type of such composite sequences, which compensate for pulse area errors, relative phase errors of over 10% can be tolerated with reasonably short sequences while maintaining the fidelity above the 99.99% quantum computing benchmark. In the second type of composite sequences, which compensate for simultaneous pulse area and detuning errors, relative phase errors of over 5% can be compensated.
Reducing errors in quantum gates is critical to the development of quantum computers. To do so, any distortions in the control signals should be identified; however, conventional tools are not always applicable when part of the system is under high vacuum, cryogenic, or microscopic. Here, we demonstrate a method to detect and compensate for amplitude-dependent phase changes, using the qubit itself as a probe. The technique is implemented using a microwave-driven trapped-ion qubit, where correcting phase distortions leads to a threefold improvement in the error of single-qubit gates implemented with pulses of different amplitudes, to attain state-of-the-art performance benchmarked at 1.6 ( 4 ) × 10 − 6 error per Clifford gate. Published by the American Physical Society 2024