Quantum random number generation using spatial quantum noise of light

Traditional random number generators often rely on deterministic algorithms (pseudo-random) or macroscopic physical processes that can be theoretically predicted. This study investigates whether the intrinsic spatial quantum fluctuations of light, captured via an electron-multiplying charge-coupled device (EMCCD), can serve as a robust, high-speed source for true quantum random number generation (QRNG) without requiring complex cryptographic extraction algorithms.

The researchers utilized a coherent laser source split into two balanced beams, which were then projected onto an EMCCD camera. By operating the camera in a high-speed kinetic mode, they captured the spatial intensity fluctuations of these light states. The team employed a frame-subtraction technique to isolate the pure quantum shot noise from static classical background noise. They verified the quantum nature of this noise through spatial cross-correlation and auto-correlation analyses, ensuring that each pixel acted as a statistically independent entropy source. Finally, they extracted the three least significant bits (LSBs) from the digitized intensity values to generate a raw bitstream, which was then validated using the NIST SP 800-22 and Marsaglia Diehard statistical test suites.

The study demonstrates that spatial quantum noise is a highly effective entropy source. The system achieved an instantaneous random bit generation rate of 5.92 Gbps, with the potential to scale up to 11.45 Gbps by utilizing the full sensor area. Because the variance of the quantum-limited fluctuations significantly exceeded the digitization step size, the raw bits were sufficiently random to pass all standard statistical tests without the need for additional entropy extractors. However, the sustained throughput is currently bottlenecked by the 1 MHz serial readout speed of the EMCCD, limiting continuous output to 7.5 Mbps.