Entangling Qubits with High Fidelity via Nonlocal Echo Sequences

University Of Maryland Baltimore County, Baltimore MD

Investigators

Abstract

Quantum computers have the potential to provide numerous benefits for the modern information-based economy, for example faster search algorithms, enhanced security, and more efficient ways to do mathematical calculations. However, it is challenging to make reliable components, such as quantum gates, for quantum computers. A quantum computer operates quite differently from current computers, since it takes advantage of the fact that an unusual set of rules, quantum mechanics, governs the behavior of the very small objects that make up the logical bits, or "qubits" of the computer. The problem is that qubits are very delicate, and any imperfections in the control or undesired perturbations (noise) can ruin the computation. The goal of this project is to discover ways to operate a quantum computer so that it self corrects certain types of errors that may occur during the computation. This will be done through the use of a controlled sequence of light pulses to "entangle" pairs of qubits, a unique feature of quantum mechanics that allows them to share information in such a way that it cannot be accessed by measuring the qubits individually. By designing a sequence of control signals that can reliably entangle pairs of qubits, it should be possible to enable qubits to undergo a self-correcting trajectory and end up in a particular, desired quantum state even in the presence of noise, bringing the field of quantum computation a step closer to realization. This project will produce a novel two-qubit dynamically protected entangling protocol that assumes only access to high-fidelity single-qubit gates (which can be produced with existing methods) and a noisy, but well-characterized, two-qubit gate. The primary application of this project will be to the entangling of quantum dot spin qubits, since noise in these otherwise very promising systems is particularly limiting at the moment. The noise (both electrical and hyperfine) is predominantly low-frequency, typically with a 1/f power spectral density, so that such a composite pulse sequence approach is reasonable. A combination of analytical and computational methods will be used to search for optimally robust sequences, using simulated randomized benchmarking to characterize their performance. The solutions will be designed to be modular such that they can be used simultaneously with pulse shaping and other fidelity-enhancing techniques to substantially reduce errors within the logical space of any qubit platform. This is a vital step towards the application of scalable, fault-tolerant quantum computing.