Quantum Computing and Sensing: Engineers Show a Quantum Advantage

Researchers at the University of Arizona College of Engineering and James C. Wyant College of Optical Sciences have experimentally demonstrated that quantum resources can improve technology today.

Quantum computing, and quantum sensing, have the potential to be far more powerful than their traditional counterparts. A fully functioning quantum computer could solve equations in seconds, whereas a classical computer would take thousands of years. It could also profoundly impact areas such as biomedical imaging and autonomous driving.

But technology is still far from perfect.

In reality, despite the widespread belief in quantum technology’s impact, only a few researchers have been capable of proving that quantum methods are superior to their traditional counterparts.

Researchers from the University of Arizona have shown that quantum computing has an advantage over traditional systems in a paper published in the journal Physical Review.

“Demonstrating a quantum benefit is a long-sought goal in the community, and very few experiments have been able to demonstrate it,” stated Zheshen Zhang (assistant professor of materials science and engineers, principal investigator at the UArizona Quantum Information and Materials Group and co-author of the paper). “We want to show how quantum technology can benefit real-world applications.”

Quantum: How and When it Works

Quantum computing and other quantum processes are based on small, powerful units called qubits. Classic computers use bits of information, which can exist in either 0s or 1s. However, qubits can exist simultaneously in both. They are both fragile and powerful because of this duality. These delicate qubits can collapse at any time, so it is essential to use error correction. This corrects errors as they occur.

Assistant professors in the College of Engineering are Zheshen Zhang (left), PI of Quantum Information Theory Group, and Quntao Zhuang, PI of Quantum Information and Materials Group. Credit: University of Arizona

John Preskill, a well-known physicist at the California Institute of Technology, describes the quantum field as “noisy intermediate quantum” or NISQ. In the NISQ era, quantum computers can perform tasks requiring only 50 to a few hundred qubits. However, there is a lot of noise or interference. Anything more than that and the noise will overwhelm the usefulness, leading to everything falling apart. It would take 10,000 to several millions of qubits to perform practically practical quantum applications.

Imagine creating a system that ensures every meal you prepare turns out perfect and giving it to children who don’t have the right ingredients. It will be a great system when the children are older and have access to all the necessary ingredients. The system’s utility is limited until then. Quantum computations will also be limited until the error correction is developed, which can lower noise levels.

Advantages of Entanglement

The paper describes a combination of quantum and classical techniques in an experiment. It used three sensors to classify radiofrequency signals’ average amplitude, angle, and magnitude.

Entanglement is a quantum resource that allows sensors to exchange information. It has two main benefits. First, it increases sensitivity and decreases errors. Entanglement enables the sensors to evaluate global properties, not just specific system parts. This is useful in applications that only require a binary answer. For example, medical imaging only requires researchers to know some of the cells in a tissue sample. They need to know if there is one cancerous cell. This same principle applies to the detection of hazardous chemicals in drinking water.

The experiment showed that quantum entanglement provided a slight advantage to the sensors over the classical ones and reduced the chance of errors by a significant margin.

Quntao Zhuang is an assistant professor in electrical and computer engineering and the principal investigator at the Quantum Information Theory Group. He said that the idea of using entanglement as a way to improve sensors does not have to be limited to one type of sensor. Consider lidar (Light Detection & Ranging) as an application for self-driving cars.

Zhang and Zhuang developed the theory for the experiment and published it in a 2019 Physical Review X article. The new paper was co-authored by Yi Xia (a doctoral student at the James C. Wyant College of Optical Sciences) and Wei Li (a postdoctoral researcher specializing in materials science, engineering, and research).

Qubit Classifiers

Existing applications use a combination of classical and quantum processing in the NISQ age. Still, they rely upon preexisting classic datasets that must then be converted and classified into the quantum realm. Imagine taking photos of cats or dogs and then uploading them into a system that uses quantum methods for labeling the photos as “cat”/”dog.”

The team uses quantum sensors to collect data and tackles the labeling problem from a new angle. This is more like using a quantum camera to label the photos as “dog” or “cat,” depending on how they are taken.

Zhuang stated that many algorithms take data from a computer disk and convert it into a quantum system. This takes time and effort. Our system evaluates physical processes in real-time to solve a different problems.

They are excited about the potential future applications of their work in quantum computing and quantum sensing. The team envisions one day integrating all their experimental setups on a chip dipped in a biomaterial to detect disease or harmful chemicals.

Zhang stated it was a new paradigm in quantum computing, machine learning, and sensors. It creates a bridge between all these domains.