Progress Report
Last updated:
Scalable and Robust Integrated Quantum Communication System[1] Development of new network architectures and protocols to realize robust and large-scale communication networks
Progress until FY2025
1. Outline of the project
The performance of quantum computers will continue to improve, but at any given point in time there is a state-of-the-art performance limit. Distributed quantum computing is a mechanism for going beyond that limit by interconnecting multiple quantum computers and using them together as a larger single computational system. To realize this, it is not sufficient simply to connect many quantum computers. It is essential to build a distributed quantum information processing environment in which the computational capabilities of individual quantum computers can be integrated and used to execute a single quantum application as a whole.
To establish such an environment, this R&D item has investigated the design of distributed quantum computer architectures, based on hardware performance data and theoretical constraints provided by the other R&D items in this project. When the project began, research on distributed quantum computer architectures was still limited worldwide, and the design space was extremely large. The outcomes of this R&D item have provided a basis for identifying a narrower set of promising architectural designs.
In particular, obtaining basic design principles for an efficient and scalable quantum network that can handle large amounts of quantum data without performance degradation caused by contention for computational resources represents a significant step toward the industrialization of distributed quantum computing systems. Our research proposing a quantum computer network protocol for resource management received an award at IEEE Quantum Week, the world’s largest international conference in quantum information engineering.
2. Outcome so far
As a basic design principle for the quantum network described above, we identified the Q-Fly architecture (Fig. 1) and developed the designs and specifications required to implement its basic group structure (red dashed circle in Fig. 1). To realize this quantum network system, we developed a Time-to-Digital Converter (TDC) used for quantum network control (Fig. 2) and device-control software, PnPQ, and released them as open-source software (Fig. 3). We also provided these technologies to the other R&D items and successfully implemented a prototype quantum network system in which multiple end nodes operate cooperatively under a network protocol, while photons are routed through optical switching to establish entanglement between selected quantum nodes. We then evaluated the performance of this prototype system.
In addition, we designed a silicon photonics chip suitable for quantum optical switching and obtained a patent for this technology (Fig. 4).We also conducted research to identify efficient quantum communication memory management methods and corresponding quantum link architectures for distributed quantum systems that implement quantum error-correcting codes and consume large amounts of entanglement. Through simulation, we confirmed that the Memory-Source-Memory link architecture achieves high performance for short-distance communication within a data center, such as over a distance of 10 meters. We also proposed “Entanglement Ejection”, a method that enables resource-efficient operation even in relatively noisy environments. Furthermore, we developed a method for supporting quantum error-correcting codes in distributed quantum systems by aggregating multiple communication links, in preparation for cases where the bandwidth of a single link is insufficient.
As an additional outcome, we developed an entanglement error-detection method that can also provide diagnostic information on abnormalities in quantum links. This eliminates the need to consume entanglement, which is an expensive computational resource, solely for quantum network management, making it possible to use all available entanglement for application execution.
Furthermore, we developed quantum network simulators and evaluation software for estimating the performance of distributed quantum computing systems, thereby laying the groundwork for accelerating future research on distributed quantum computers. We also confirmed that the performance of hardware components can be simulated in detail, enabling the simulator to be used for network design, improvement, and performance prediction.
3. Future plans
We expect that the technologies developed in this R&D item, including Q-Fly, are particularly compatible with optical technologies and can be applied to any physical platform that uses photons for quantum communication. Using the technologies developed in this project, we will implement and evaluate a prototype system that routes entangled photons emitted from quantum computers located at spatially separated sites over the Q-Fly topology, thereby generating remote entanglement between arbitrary qubits. Through this work, we will further strengthen the pathway toward building and industrializing large-scale, efficient distributed quantum computers.