Quantum Interference Enables Constant-Time Quantum Information Processing

Facebook X LinkedIn Email
Quantum interference can be used to process large data sets quickly and accurately, according to a study done by the University of Warsaw, in collaboration with the University of Oxford and NIST. The researchers showed that a simple quantum gate, consisting of an optical device such as a beamsplitter, can be used to compute the Kravchuk transform — an approach that could be better suited to digital image processing and computer vision than the commonly used fast Fourier transform (FFT) algorithm.

There currently exists no fast algorithm to compute the Kravchuk transform, but the scientists circumvented this limitation using quantum mechanics. The computation they devised is a one-step operation that can be implemented with a single gate.

The researchers used the quantum gate to interfere two quantum states, dividing photons between two outputs. When two states of quantum light entered the input ports of the device from two sides, the states interfered. For example, when two identical photons entered the device at the same time they would bunch into a pair and exit together by the same port. This effect, known as the Hong-Ou-Mandel effect, can extend to states made up of many particles, the researchers said. By interfering “packets” of many indistinguishable photons, which encoded the information, the researchers achieved a specialized quantum computer that could compute the Kravchuk transform.

The experiment was performed at the University of Oxford, where a special setup was built to produce multiphoton quantum states. The laboratory was equipped with transmission edge sensors, developed by NIST, which counted photons to allow a precise reading of the quantum state leaving the beamsplitter and, thus, a precise reading of the computation results. The researchers said that a computation of the quantum Kravchuk transform always takes the same time, regardless of the size of the input data set.

To obtain these results in practice, one needs to perform the experiment several hundred times to get the statistics. However, this does not take long, the team said, because the laser produces dozens of millions of multiphoton “packets” per second.

Science, medicine, engineering, and information technology require efficient processing of data such as images, sound, radio signals, and records coming from various probes and cameras. Since the 1970s, this has been achieved by means of the FFT algorithm. According to the researchers, the Kravchuk transform performs better than the FFT in processing deformed or noisy data.

The use of quantum interference to perform a computation of a quantum Kravchuk transform could find applications in the development of new quantum technologies and quantum algorithms. Its range of uses could go beyond quantum photonics, since a similar quantum interference can be observed in many different types of quantum systems, the team said.

The University of Warsaw has applied for an international patent for this innovation. The scientists hope that the Kravchuk transform will soon find use in quantum computation, where it could become a component of new algorithms, especially in hybrid quantum-classical computers that merge quantum circuits with conventional digital layouts.

The research was published in Science Advances ( 

Quantum interference in service of information technology. Courtesy of M. Czerniawski, L. Kaluza, Promotion Office UW.

Published: July 2019
quantum optics
The area of optics in which quantum theory is used to describe light in discrete units or "quanta" of energy known as photons. First observed by Albert Einstein's photoelectric effect, this particle description of light is the foundation for describing the transfer of energy (i.e. absorption and emission) in light matter interaction.
Research & TechnologyeducationEuropeUniversity of WarsawAmericasNISTUniversity of Oxfordquantum opticsquantum photonicsquantum gatesingle photonsKravchuk transformquantum interferencequantum information processing

We use cookies to improve user experience and analyze our website traffic as stated in our Privacy Policy. By using this website, you agree to the use of cookies unless you have disabled them.