Project

Back to overview

Noise-biased qubits for hardware-efficient quantum error detection

Applicant Grimm Alexander
Number 197255
Funding scheme Project funding
Research institution Paul Scherrer Institut
Institution of higher education Paul Scherrer Institute - PSI
Main discipline Condensed Matter Physics
Start/End 01.09.2021 - 31.08.2025
Approved amount 328'964.00
Show all

All Disciplines (2)

Discipline
Condensed Matter Physics
Electrical Engineering

Keywords (3)

Quantum science and technology; Quantum computation; Superconducting circuits

Lay Summary (German)

Lead
Alexander Grimm
Lay summary

Das Ziel dieses Projekts ist es ein System für die robuste Detektion von Fehlern in supraleitenden Quantenbits zu entwickeln. Dabei handelt es sich um einen Schaltkreis der solche Fehler nicht nur erkennen kann, sondern auch zu verhindert, dass durch die Messung weitere Fehler verursacht werden.

Quantenbits, die grundlegenden Einheiten der Quanteninformatik entsprechen der Information, die in zwei Zustände eines Quantensystems gespeichert werden kann. Diese Zustände sind üblicherweise sehr empfindlich und verlieren ihre Quanteneigenschaften mit der Zeit. Das führt zu Fehlern in der gespeicherten Information. Quantenfehlerkorrektur zielt darauf ab Quanteninformationen redundant abzuspeichern in einer Art, die es möglich macht solche Fehler durch Messungen aufzuspüren und zu beheben. Dabei muss besonderes Augenmerk darauf gelegt werden, dass diese Messung nicht weitere Fehler verursacht. Das kann im Regelfall nicht garantiert werden, denn generell ist der Messapparat selbst fehlerbehaftet. Hier werden wir in neuartiges Quantensystem, das auf Überlagerungen von Oszillationen in einem supraleitenden Schaltkreis basiert, für diese Messung verwenden. Der Vorteil dieses Systems ist, dass es selbst gegen die Fehler, die für das gemessene Qubit am schädlichsten sind, immun ist.

Direct link to Lay Summary Last update: 28.06.2021

Responsible applicant and co-applicants

Employees

Name Institute

Abstract

Quantum information processing has now reached a stage where we are building quantum machines that can no longer be efficiently described by classical computers. Further progress in the field could lead to the development of quantum computers that can solve certain problems much faster than their classical counterparts. Even before this goal is reached, the first other useful devices such as quantum simulators could be built.To a large extend, progress towards these goals will depend on our ability to store and control quantum information in more and more complex systems as we scale up the size of our quantum machines. At this point, we do not yet know if there is an intrinsic limit to this process, but the field of quantum error correction (QEC) provides us with a recipe for how to explore this fundamentally and practically interesting question. QEC is based on the following basic premises: If a bit of quantum information, or “qubit”, is encoded into a physical quantum two-level system, it always suffers from errors due to decoherence. The same information can be protected against errors if it is redundantly encoded into a multi-level quantum system in such a way that the effects of decoherence can be detected and corrected. The resulting effective qubit is referred to as a “logical qubit”. A crucial part of this process is the extraction of information about errors (error syndromes) with the help of an ancillary quantum system. This creates however a predicament: in general, this ancilla is a quantum-two-level system which itself suffers from decoherence. Consequently, the coupling to this system exposes the logical qubit to additional errors. In order for QEC to be feasible, ancilla errors must not propagate to the logical qubit as additional uncorrectable errors. The current schemes for achieving this “fault tolerance” to ancilla errors come at a heavy cost of increased hardware overhead such as the use of multiple ancilla systems. Here, I propose to implement a hardware-effcient approach to this challenge by using a so-called “noise-biased” qubit as an ancilla. Noise bias refers to the strong suppression of one type of quantum error. This can be exploited by engineering the coupling of the ancilla to the logical qubit such that only the suppressed error would propagate, making the error syndrome measurement intrinsically fault-tolerant. In practice, our approach is to encode such a qubit into superpositions of two opposite-phase coherent states in a nonlinear superconducting resonator. In this proposal, I intend to experimentally demonstrate the utility of this system, sometimes called a Schrödinger cat qubit, as an ancilla in QEC. We will perform error detection measurements for various QEC approaches and implement this system in a way that is compatible with leading initiatives for superconducting quantum computation.
-