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:1810.07184 · Physical review. B./Physical review. B

Conditions allowing error correction in driven qubits

Robert E. Throckmorton, S. Das Sarma

Published Jan 17, 2019• 9 citations
arXivHTMLPDF

Abstract

We consider a qubit that is driven along its logical $z$ axis, with noise along the $z$ axis in the driving field $\mathrm{\ensuremath{\Omega}}$ proportional to some function $f(\mathrm{\ensuremath{\Omega}})$, as well as noise along the logical $x$ axis. We establish that whether or not errors due to both types of noise can be canceled out, even approximately, depends on the explicit functional form of $f(\mathrm{\ensuremath{\Omega}})$ by considering a power-law form, $f(\mathrm{\ensuremath{\Omega}})\ensuremath{\propto}{\mathrm{\ensuremath{\Omega}}}^{k}$. In particular, we show that such cancellation is impossible for $k=0,\phantom{\rule{0.16em}{0ex}}1$, or any even integer. However, any other odd integer value of $k$ besides 1 does permit cancellation; in fact, we show that both types of errors can be corrected with a sequence of four square pulses of equal duration. We provide sets of parameters that correct for errors for various rotations and evaluate the error, measured by the infidelity, for the corrected rotations versus the na\"{\i}ve rotations, i.e., the operations that, in the complete absence of noise, would produce the desired rotations (in this case a single pulse of appropriate duration and magnitude). We also consider a train of four trapezoidal pulses, which take into account the fact that there will be, in real experimental systems, a finite rise time, again providing parameters for error-corrected rotations that employ such pulse sequences. Our dynamical decoupling error correction scheme works for any qubit platform as long as the errors are quasistatic.

DOI
10.1103/physrevb.99.045422

Related papers

Qubit noise spectroscopy using a continuous driving field

Daniel K. Park, Jonathan Baugh

The optimization of dynamical decoupling and quantum error correction for a particular qubit realization is based on a detailed knowledge of the noise properties. Spectroscopy of single-axis noise using dynamical decoupling pulse sequences has garnered much recent attention. Here we consider noise spectroscopy based on a spin-locking type pulse sequence, i.e. a continuous-wave (CW) on-resonance driving field. We show that a heuristic filter function approach produces a qualitatively correct (but quantitatively incorrect) result, whereas a 0th-order average Hamiltonian calculation is shown to agree with the result predicted by the generalized Bloch equations. We further calculate up to 2nd-order average Hamiltonian corrections and show the deviation from the generalized Bloch equation result. This shows that noise spectroscopy using continuous fields, in some cases simpler to implement and more robust to errors than pulsed schemes, can be rigorously analyzed and criteria for reliable measurements can be established. Finally, a solid-state nuclear magnetic resonance experiment is presented which demonstrates that the CW and pulsed methods agree within experimental error. The noise, due to magnetization fluctuations in a dipolar coupled proton spin bath, is found to obey a roughly 1/ω power law decay in the range of frequencies ω investigated.

arXiv:1308.6310•Aug 28, 2013•0 citations

Efficiency of dynamical decoupling sequences in the presence of pulse errors

Zhihao Xiao, Lewei He, Wen-ge Wang

For a generic dynamical decoupling sequence employing a single-axis control, we study its efficiency in the presence of small errors in the direction of the controlling pulses. In the case that the corresponding ideal dynamical decoupling sequence produces sufficiently good results, the impact of the errors is found to scale as $\ensuremath{\xi}{}^{2}$, with negligible first-order effect, where $\ensuremath{\xi}$ is the dispersion of the random errors. This analytical prediction is numerically tested in a model in which the environment is modeled by one qubit coupled to a quantum kicked rotator in chaotic motion. In this model, with periodic pulses applied to the qubit in the environment, it is found numerically that periodic bang-bang control may outperform Uhrig dynamical decoupling.

arXiv:1101.5430•Physical Review A•Mar 30, 2011•14 citations

Crosstalk error correction through dynamical decoupling of single-qubit gates in capacitively coupled singlet-triplet semiconductor spin qubits

Donovan L. Buterakos, Robert E. Throckmorton, S. Das Sarma

In addition to magnetic field and electric charge noise adversely affecting spin-qubit operations, performing single-qubit gates on one of multiple coupled singlet-triplet qubits presents a new challenge: crosstalk, which is inevitable (and must be minimized) in any multiqubit quantum computing architecture. We develop a set of dynamically corrected pulse sequences that are designed to cancel the effects of both types of noise (i.e., field and charge) as well as crosstalk to leading order, and provide parameters for these corrected sequences for all 24 of the single-qubit Clifford gates. We then provide an estimate of the error as a function of the noise and capacitive coupling to compare the fidelity of our corrected gates to their uncorrected versions. Dynamical error correction protocols presented in this work are important for the next generation of singlet-triplet qubit devices where coupling among many qubits will become relevant.

arXiv:1711.02092•Physical review. B./Physical review. B•Jan 29, 2018•38 citations

Dynamical Decoupling in the Presence of Realistic Pulse Errors

Alexei M. Tyryshkin, Zhi‐Hui Wang, Wenxian Zhang +4 more

One of the most significant hurdles to be overcome on the path to practical quantum information processors is dealing with quantum errors. Dynamical decoupling is a particularly promising approach that complements conventional quantum error correction by eliminating some correlated errors without the overhead of additional qubits. In practice, the control pulses used for decoupling are imperfect and thus introduce errors which can accumulate after many pulses. These instrumental errors can destroy the quantum state. Here we examine several dynamical decoupling sequences, and their concatenated variants, using electron spin resonance of donor electron spins in a $^{28}$Si crystal. All of the sequences cancel phase noise arising from slowly fluctuating magnetic fields in our spectrometer, but only those sequences based upon alternating $π$-rotations about the X- and Y-axes in the rotating frame (XYXY sequences) demonstrate the ability to store an arbitrary quantum state. By comparing the experimental results with a detailed theoretical analysis we demonstrate that the superior performance of XYXY sequences arises from the fact that they are self-correcting for the dominant instrumental pulse errors in magnetic resonance experiments. We further find that concatenated sequences perform better than the periodic variants, maintaining near 100% fidelities for spin states even after several hundred control pulses. Intuitively, one would expect the instrumental error to increase with the number of pulses in the sequence but we show that the dominant first-order error does not increase when concatenating the XYXY sequence.

arXiv:1011.1903•Nov 8, 2010•27 citations

Qubit decoherence under two-axis coupling to low-frequency noises

Guy Ramon, Łukasz Cywiński

Many solid-state qubit systems are afflicted by low-frequency noise mechanisms that operate along two perpendicular axes of the Bloch sphere. Depending on the qubit control fields, either noise can be longitudinal or transverse to the quantization axis of the qubit, thus affecting its dynamics in distinct ways, generally contributing to decoherence that goes beyond pure dephasing. Here, we present a theory that provides a unified platform to study dynamics of a qubit subjected to two perpendicular low-frequency noises (assumed to be Gaussian and uncorrelated) under dynamical decoupling pulse sequences. The theory is demonstrated by the commonly encountered case of power law noise spectra, where approximate analytical results can be obtained.

arXiv:2112.03438•Physical review. B./Physical review. B•Jan 31, 2022•5 citations

Fastest pulses that implement dynamically corrected single-qubit phase gates

Junkai Zeng, Edwin Barnes

Dynamically correcting for unwanted interactions between a quantum system and its environment is vital to achieving the high-fidelity quantum control necessary for a broad range of quantum information technologies. In recent work, we uncovered the complete solution space of all possible driving fields that suppress transverse quasistatic noise errors while performing single-qubit phase gates. This solution space lives within a simple geometrical framework that makes it possible to obtain globally optimal pulses subject to a set of experimental constraints by solving certain geometrical optimization problems. In this work, we solve such a geometrical optimization problem to find the fastest possible pulses that implement single-qubit phase gates while canceling transverse quasistatic noise to second order. Because the time-optimized pulses are not smooth, we provide a method based on our geometrical approach to obtain experimentally feasible smooth pulses that approximate the time-optimal ones with minimal loss in gate speed. We show that in the presence of realistic constraints on pulse rise times, our smooth pulses significantly outperform sequences based on ideal pulse shapes, highlighting the benefits of building experimental waveform constraints directly into dynamically corrected gate designs.

arXiv:1802.10241•Physical Review A•Jul 2, 2018•52 citations
We consider a qubit that is driven along its logical zzz axis, with noise along the zzz axis in the driving field \ensuremathΩ\mathrm{\ensuremath{\Omega}}\ensuremathΩ proportional to some function f(\ensuremathΩ)f(\mathrm{\ensuremath{\Omega}})f(\ensuremathΩ), as well as noise along the logical xxx axis. We establish that whether or not errors due to both types of noise can be canceled out, even approximately, depends on the explicit functional form of f(\ensuremathΩ)f(\mathrm{\ensuremath{\Omega}})f(\ensuremathΩ) by considering a power-law form, f(\ensuremathΩ)\ensuremath∝\ensuremathΩkf(\mathrm{\ensuremath{\Omega}})\ensuremath{\propto}{\mathrm{\ensuremath{\Omega}}}^{k}f(\ensuremathΩ)\ensuremath∝\ensuremathΩk. In particular, we show that such cancellation is impossible for k=0,1k=0,\phantom{\rule{0.16em}{0ex}}1k=0,1, or any even integer. However, any other odd integer value of kkk besides 1 does permit cancellation; in fact, we show that both types of errors can be corrected with a sequence of four square pulses of equal duration. We provide sets of parameters that correct for errors for various rotations and evaluate the error, measured by the infidelity, for the corrected rotations versus the na"{\i}ve rotations, i.e., the operations that, in the complete absence of noise, would produce the desired rotations (in this case a single pulse of appropriate duration and magnitude). We also consider a train of four trapezoidal pulses, which take into account the fact that there will be, in real experimental systems, a finite rise time, again providing parameters for error-corrected rotations that employ such pulse sequences. Our dynamical decoupling error correction scheme works for any qubit platform as long as the errors are quasistatic.
The optimization of dynamical decoupling and quantum error correction for a particular qubit realization is based on a detailed knowledge of the noise properties. Spectroscopy of single-axis noise using dynamical decoupling pulse sequences has garnered much recent attention. Here we consider noise spectroscopy based on a spin-locking type pulse sequence, i.e. a continuous-wave (CW) on-resonance driving field. We show that a heuristic filter function approach produces a qualitatively correct (but quantitatively incorrect) result, whereas a 0th-order average Hamiltonian calculation is shown to agree with the result predicted by the generalized Bloch equations. We further calculate up to 2nd-order average Hamiltonian corrections and show the deviation from the generalized Bloch equation result. This shows that noise spectroscopy using continuous fields, in some cases simpler to implement and more robust to errors than pulsed schemes, can be rigorously analyzed and criteria for reliable measurements can be established. Finally, a solid-state nuclear magnetic resonance experiment is presented which demonstrates that the CW and pulsed methods agree within experimental error. The noise, due to magnetization fluctuations in a dipolar coupled proton spin bath, is found to obey a roughly 1/ω power law decay in the range of frequencies ω investigated.
For a generic dynamical decoupling sequence employing a single-axis control, we study its efficiency in the presence of small errors in the direction of the controlling pulses. In the case that the corresponding ideal dynamical decoupling sequence produces sufficiently good results, the impact of the errors is found to scale as $\ensuremath{\xi}{}^{2}$, with negligible first-order effect, where $\ensuremath{\xi}$ is the dispersion of the random errors. This analytical prediction is numerically tested in a model in which the environment is modeled by one qubit coupled to a quantum kicked rotator in chaotic motion. In this model, with periodic pulses applied to the qubit in the environment, it is found numerically that periodic bang-bang control may outperform Uhrig dynamical decoupling.
In addition to magnetic field and electric charge noise adversely affecting spin-qubit operations, performing single-qubit gates on one of multiple coupled singlet-triplet qubits presents a new challenge: crosstalk, which is inevitable (and must be minimized) in any multiqubit quantum computing architecture. We develop a set of dynamically corrected pulse sequences that are designed to cancel the effects of both types of noise (i.e., field and charge) as well as crosstalk to leading order, and provide parameters for these corrected sequences for all 24 of the single-qubit Clifford gates. We then provide an estimate of the error as a function of the noise and capacitive coupling to compare the fidelity of our corrected gates to their uncorrected versions. Dynamical error correction protocols presented in this work are important for the next generation of singlet-triplet qubit devices where coupling among many qubits will become relevant.
One of the most significant hurdles to be overcome on the path to practical quantum information processors is dealing with quantum errors. Dynamical decoupling is a particularly promising approach that complements conventional quantum error correction by eliminating some correlated errors without the overhead of additional qubits. In practice, the control pulses used for decoupling are imperfect and thus introduce errors which can accumulate after many pulses. These instrumental errors can destroy the quantum state. Here we examine several dynamical decoupling sequences, and their concatenated variants, using electron spin resonance of donor electron spins in a 28^{28}28Si crystal. All of the sequences cancel phase noise arising from slowly fluctuating magnetic fields in our spectrometer, but only those sequences based upon alternating πππ-rotations about the X- and Y-axes in the rotating frame (XYXY sequences) demonstrate the ability to store an arbitrary quantum state. By comparing the experimental results with a detailed theoretical analysis we demonstrate that the superior performance of XYXY sequences arises from the fact that they are self-correcting for the dominant instrumental pulse errors in magnetic resonance experiments. We further find that concatenated sequences perform better than the periodic variants, maintaining near 100% fidelities for spin states even after several hundred control pulses. Intuitively, one would expect the instrumental error to increase with the number of pulses in the sequence but we show that the dominant first-order error does not increase when concatenating the XYXY sequence.
Many solid-state qubit systems are afflicted by low-frequency noise mechanisms that operate along two perpendicular axes of the Bloch sphere. Depending on the qubit control fields, either noise can be longitudinal or transverse to the quantization axis of the qubit, thus affecting its dynamics in distinct ways, generally contributing to decoherence that goes beyond pure dephasing. Here, we present a theory that provides a unified platform to study dynamics of a qubit subjected to two perpendicular low-frequency noises (assumed to be Gaussian and uncorrelated) under dynamical decoupling pulse sequences. The theory is demonstrated by the commonly encountered case of power law noise spectra, where approximate analytical results can be obtained.
Dynamically correcting for unwanted interactions between a quantum system and its environment is vital to achieving the high-fidelity quantum control necessary for a broad range of quantum information technologies. In recent work, we uncovered the complete solution space of all possible driving fields that suppress transverse quasistatic noise errors while performing single-qubit phase gates. This solution space lives within a simple geometrical framework that makes it possible to obtain globally optimal pulses subject to a set of experimental constraints by solving certain geometrical optimization problems. In this work, we solve such a geometrical optimization problem to find the fastest possible pulses that implement single-qubit phase gates while canceling transverse quasistatic noise to second order. Because the time-optimized pulses are not smooth, we provide a method based on our geometrical approach to obtain experimentally feasible smooth pulses that approximate the time-optimal ones with minimal loss in gate speed. We show that in the presence of realistic constraints on pulse rise times, our smooth pulses significantly outperform sequences based on ideal pulse shapes, highlighting the benefits of building experimental waveform constraints directly into dynamically corrected gate designs.