NEC: First Unit Cell for Scaling up to a Fully-connected Quantum Annealing Architecture

Print Friendly, PDF & Email

Tokyo, March 17, 2022 – NEC Corporation (NEC; TSE: 6701) has developed the world’s first LHZ scheme (*1) unit cell facilitating scaling up to a fully-connected architecture using superconducting parametron (*2) quantum bits (qubits). NEC has succeeded in demonstrating quantum annealing operations using these qubits, which will enable high accuracy calculations. Via this achievement, NEC has made progress toward the production of a quantum annealing machine (*3), which is a type of quantum computer.

Figure 1: LHZ scheme unit cell (left) and diagram of scaled up architecture (right)

Combinational optimization is important for finding solutions to complex social issues. It aims to find the optimal solution from an enormous set of potential choices. In 1999, NEC developed a superconducting qubit for use in gate-type quantum computers. Since then, NEC has applied the technology in the research and development of a quantum annealing machine using superconducting parametron qubits that can solve combinatorial optimization problems at high speed and with great accuracy.

NEC has developed a four qubit unit cell of the LHZ scheme. This enables scaling to multiple fully-connected logical qubits utilizing superconducting parametron and circuit coupling technology. NEC has achieved a world first by successfully solving small-scale combinatorial optimization problems via quantum annealing using this new technology. In another world first, NEC has also developed a three-dimensional structure technology that efficiently connects many LHZ scheme unit cells arranged in a tile pattern with external devices.

Figure 2: Diagram of the three-dimensional structure technology

By replicating the unit cell in a tile pattern, it is possible to easily create a structure where many qubits are logically connected to each other, whilst maintaining the features of the superconducting parametron that allow it to perform calculations with high accuracy. NEC has made progress toward the realization of a quantum annealing machine that can solve large-scale and complex combinatorial optimization problems at high speed.

NEC is working to develop a quantum annealing machine using superconducting parametrons as a project (*4) commissioned by the New Energy and Industrial Technology Development Organization (NEDO). NEC is currently conducting research and development to improve the integration of superconducting parametrons in a fully connected architecture, with the aim of realizing quantum annealing machines by 2023. NEC will use these results as building blocks to further accelerate the development of quantum computers.


  • (*1)
    LHZ scheme: LHZ is an abbreviation for a technique proposed by Lechner, Hauke and Zoller. As the number of qubits increases, it becomes difficult in hardware to directly connect each qubit to each of the other qubits. To solve this problem, ParityQC proposed with LHZ a transformation that enables fully connected qubits to be obtained using qubits that are physically connected to only their nearest neighbours. A unit cell, composed of four qubits and a central coupling circuit connecting them can be replicated using a tile-like pattern.
    About ParityQC: new window
  • (*2)
    Superconducting parametron: A superconducting resonant circuit composed of Josephson junctions and capacitors which oscillates with different phases and can be used as a qubit. The lifetime of the qubit (which determines the upper limit of the time for which high-speed operation is possible) is an order of magnitude longer than that of magnetic flux qubits. It is expected that calculations performed within a fixed time period will have improved accuracy.
  • (*3)
    Quantum Annealing Machine: A computer that exploits the laws of quantum mechanics to search for the minimum energy state of a cost function. The minimum energy state corresponds to the solution of the combinatorial optimization problem. The smallest unit to be calculated is a qubit. As the number of qubits increases and the connectivity between the qubits increases, larger and more complex combinatorial optimization problems can be solved.
  • (*4)
    Project for Innovative AI Chip and Next-Generation Computing Technology Development
    new window