Using the time-domain multiplexing techniques developed by the R&D theme 1, we have generated large-scale cluster states that are currently unachievable by other methods. In particular, the generated cluster state has a structure that allows more efficient computation compared to previously generated cluster states.
Thanks to our prior research, we have significantly improved and stabilized the laboratory environment, making it possible to construct the system using a free-space setup. This was achieved by enhancing the temperature stability of the laboratory environment, which now allows for long-term stability even in a free-space configuration. Moreover, the free-space approach offers lower losses compared to fiber-based approach. As a result, we are now able to generate higher-quality cluster states.
So far, we have successfully built the cloud-based system for the first prototype of our optical quantum computer, as shown in Figure 1. This prototype operates in a free-space setup at a clock frequency of 100 MHz and functions as an analog quantum computer with 101 input modes. The system is capable of performing programmable linear transformation on continuous variables. In addition, thanks to the time-domain multiplexing technique, it is, in principle, capable of executing operations with an arbitrary number of steps. Furthermore, we have developed the cloud system including a local server in RIKEN, a framework for remote operation of the hardware, a user authentication system, a compiler that converts quantum circuits into machine parameters, and a software development kit (SDK). These developments enable us to successfully construct a cloud-based optical quantum computer system.

In this optical quantum computer, the types and precision of measurements directly correspond to the types and precision of possible quantum operations. In the current system, only deterministic addition, subtraction, and scalar multiplication operations are implementable. However, to realize universal quantum computation, the optical quantum computer must be capable of performing “multiplication” operations, which require nonlinear measurements corresponding to deterministic nonlinear operations. We have successfully implemented a basic nonlinear measurement using the setup shown in Figure 2. In this configuration, the target quantum state is superimposed with an auxiliary quantum state, and one of the outputs is measured via homodyne detection. A nonlinear computation is then applied to the measurement result, and based on this result, the phase of the remaining quantum state is rotated before performing a second homodyne measurement. A key component of this system is high-speed digital signal processing, which we have achieved using a field-programmable gate array (FPGA) that incorporates a lookup table for the necessary computations. We are also conducting research to integrate the nonlinear operation into the optical quantum computer, aiming to realize a optical quantum computer capable of performing “multiplication” operations.