One challenge to making photon-based quantum computing practical is finding a way to manipulate and measure the thousands of qubits that are needed to process extremely large data sets. A multi-institutional research team has now demonstrated a way to map and measure large-scale photonic quantum correlation with single-photon sensitivity. In addition to the new measurement technique, which is called correlation on spatially mapped photon-level image (COSPLI), the researchers also developed a way to detect signals from single photons and their correlations in tens of millions of images. When qubits are photons, their number can be increased by increasing the number of modes encoded in the photonic degrees of freedom,that is, the polarization, frequency, time, and location measured for each photon. This increases the number of qubits without the need to use more photons, because it allows each photon to exhibit more than two states, or modes, simultaneously. The researchers had already applied this approach to the fabrication of photonic quantum chips with a state space equivalent to thousands of qubits. However, to incorporate these chips into a quantum computer, the researchers required a way to measure all the states and their photonic correlations at a single-photon level. The use of single-photon detectors to address thousands of states simultaneously would be unfeasible, the researchers said, requiring thousands of single-photon detectors at a cost of around $12 million for a single computer. After two years of work, the researchers developed methods for suppressing signal noise from individual photons, so that the single photons could be detected with each pixel of a CCD camera. Using COSPLI, the researchers were able to map photonic correlations from different states, or modes, onto the spatial mode. This enabled them to measure polarization, frequency, time, and location at the single-photon level, by allowing correlations of all the modes to be measured with the CCD camera. Experimental results of correlation measurements with Fourier transformation. Figures (a) and (b) show the corrected joint spectra of correlated photons without and with bandpass filters, respectively. Courtesy of Xian-Min Jin et al./Optica/The Optical Society. To demonstrate COSPLI, the researchers used their approach to measure the joint spectra of correlated photons in 10 million image frames. The reconstructed spectra corresponded well with theoretical calculations, thus demonstrating the reliability of the measurement and mapping method as well as the single-photon detection method. The researchers are now working to improve the imaging speed of the system from tens to millions of frames per second. “COSPLI has the potential to become a versatile solution for performing quantum particle measurements in large-scale photonic quantum computers,” said the research team leader, Xian-Min Jin, from Shanghai Jiao Tong University. “This unique approach would also be useful for quantum simulation, quantum communication, quantum sensing, and single-photon biomedical imaging. “We know it is very hard to build a practical quantum computer, and it isn’t clear yet which implementation will be the best,” Jin said. “This work adds confidence that a quantum computer based on photons may be a practical route forward.” The research was published in Optica, a publication of OSA, The Optical Society (https://doi.org/10.1364/OPTICA.6.000244).
