MIT researchers have created a quantum computing architecture that allows superconducting quantum processors to communicate in an extensible, high-fidelity manner. Quantum computers have the potential to perform tasks that are currently intractable, even on the world’s most powerful supercomputers.
Scientists are looking to use quantum computing to work like materials systems, catalyse quantum chemistry and manage challenging tasks. The development could impact areas as diverse as finance to pharmaceuticals. Harnessing this potential, however, is subject to resilient and extensible hardware.
To achieves the goals, MIT researchers addressed step one, the deterministic emission of single photons – information carriers – in a user-specified direction. Their method ensures that quantum information flows correctly more than 96% of the time. A more extensive network of interconnected quantum processors can be created, regardless of their physical separation on a computer chip.
“Quantum interconnects are an important step toward modular implementations of larger-scale machines built from smaller individual components,” explained Bharath Kannan, PhD (2022), Co-lead Author of a paper.
Kannan co-wrote the paper with Aziza Almanakly, an Electrical Engineering and Computer Science graduate student at MIT’s Research Laboratory of Electronics (RLE) in the Engineering Quantum Systems group. William D. Oliver, an MIT Professor of Electrical Engineering, Computer Science and Physics, an MIT Lincoln Laboratory Fellow, Director of the Centre for Quantum Engineering and Associate Director of RLE, is the senior author.
Quantum computing challenges
Quantum computers are entirely different from traditional computers. However, one requirement is particular: the information must be transmitted and received. The pressing problem in creating a large-scale quantum computer is linking quantum information nodes. These are smaller-scale processing nodes separated across a computer chip. As a result, standard methods of communicating electronic information cannot be mapped and employed for quantum devices.
The researchers must devise a new method to effectively interconnect the quantum information nodes carried by the photon. A photon is a light particle that carries up quantum information through waveguides. They found out to scale up data flows; the waveguide transmission must be two-way (bidirectional) to the left and right. Currently, most quantum computing systems transfer photons through only unidirectional waveguides (left or right).
Thus, Kannan and his colleagues create a bidirectional component that can support propagation in both the left and right directions and a means to choose the path at will. The ‘directional transmission’ is the first step toward bidirectional communication with much higher fidelities of quantum computing.
They used the concept to create a single physical connection that could handle any number of modules via a single waveguide. Photons can be delivered and caught by any two modules through a shared waveguide using the same module as both a transmitter and a receiver, which allows them to be scaled up.
The new architecture
Almanakly then developed a second module capable of catching that photon downstream. The researchers achieved this by creating a module with four qubits. Qubits, which are used to store and process quantum data, are the essential building elements of quantum computers.
The qubit must be entangled in a relative Bell state to be able to be directed to the left or right. The researchers choose the correct phase to move the photon to the desired direction along the waveguide. They can use the same procedure to receive the photon at another module but in reverse.
The researchers determined that their technique had a fidelity of more than 96%, which indicates that if they wanted to release a photon to the right, it did so 96% of the time. Hence, the findings would be essential to develop a modular architecture connecting several smaller-scale processors into a larger, more powerful quantum processor.