### The importance of randomness

Random number generators (RNGs) are essential in a wide range of applications, such as secure communications or stochastic simulations. To assure secure communications we encrypt data by using RNGs to generate cryptographic keys. Stochastic simulations depend on random numbers to accurately calculate financial derivatives or model precise solutions in logistics. Hence, random numbers are essential, and producing them with perfect unpredictability is a big challenge [1].

### How to generate unpredictable random numbers

In today's world, pseudo-random number generators (PRNGs) are commonly used to produce the random numbers most of the times. However, due to their algorithmic nature, pseudo-random number generators are software-based and inherently deterministic. The design of this algorithms is carefully done to eliminate statistical anomalies. However, they cannot guarantee being determined by an intelligent observer. If an observer knows the code, current state, and inputs, it can reliably predict the output.

To eliminate the strong algorithmic dependence, in practice we generate the required unpredictability by measuring physical processes. That is, randomness is a natural resource and we need to extract it from physical systems. Many processes have been used today, ranging from mouse movements of a user in a desktop computer to dedicated hardware measuring noise in standard electronics. Unfortunately, these systems are based on classical physics and therefore cannot offer any reliable solution for generating unpredictability. The reason is: classical physics is deterministic, and one cannot determine how much an observer knows. In the context of cryptography, we have to assume that an observer increases the ability to collect information with practice, time, and technology. And using inherently predictable RNGs is a risk.

### Next-generation quantum random number generators

This is where quantum comes into play. Quantum random number generators (QRNGs) create randomness by measuring quantum processes, which are, by nature fully non-deterministic. Engineering high-quality, scalable and fast quantum random number generators has been a challenge to date, and this is the trade-off we worked on at Quside to overcome. Our proprietary technology allows for fast, high-quality, and scalable production, leading to a solution that is ready for today’s unpredictability concerns and tomorrow’s performance requirements.

### Quside’s phase-diffusion quantum random number generators

Quside’s proprietary technology is based on the phase-diffusion process in semiconductor lasers. The core element of the technology is converting microscopic quantum observables, which are delicate and hard to measure, into macroscopic dynamics that are robust and easy to capture.

To do this, we modulate a semiconductor laser from below to above its threshold level or produce a stream of phase randomized optical pulses. This is called gain-switching. Then, we use an interferometer to convert the phase fluctuations into the amplitude domain, generating a stream of amplitude-randomized optical pulses at the output (see refs [2, 3] for two examples of interferometers that we use). Finally, a fast photodiode converts the photonic signal into the electronic domain, where standard electronics are used for turning the analog signal into the digital realm.

At the heart, the unpredictability of the phase-diffusion technology traces back to the process of spontaneous emission, which occurs as a result of the interaction between the quantum vacuum field and the laser’s gain medium. Quside’s technology exploits this quantum-mechanical process to produce quantum-based random numbers at multiple Gigabits per second.

### In randomness generation, quantum should mean quality.

Testing randomness is a complex matter. So the questions “how do you know it’s random?” is a hard one to answer. For this reason, since 2014 we have developed what we call the Quside^{TM} Randomness Metrology methodology [5], which allows us to place strict quality bounds on all of our devices in a transparent manner. One of the main advantages of quantum processes for randomness generation is that the core physical dynamics that underpin the process are well understood and can be modeled and tested. In many classical RNGs this is not the case, and it is unclear where the unpredictability that is observed comes from.

With our Randomness Metrology technology, all our devices have embedded capabilities to assess their unpredictability level, obtaining rigorous min-entropy bounds. This metrological capability combined with the high performance of the phase-diffusion QRNG technology proved essential in a highly demanding environment (probably the highest demanding scenario for random number generators) as was the 2015 loophole-free Bell test experiments that were reported in Nature [6] and Phys. Rev. Lett. [7, 8].

### On a journey to deploying the phase-diffusion technology at scale

Random numbers are required in all devices that need to communicate over the Internet in a secure manner. For this reason and given that there are billions of devices today in the field in a wide variety of networking configurations, scalable, chip-level technology is required. Quside’s photonic integrated quantum entropy cores enable high-quality and high-speed quantum randomness generation at a scale and performance that is unprecedented. The basics of the technology can be looked into more detail in [2, 9].

In addition to the scaling, adaptability is another key area we at Quside are committed to. There are a variety of systems that need to integrate our phase-diffusion QRNG technology, and we are making every effort to help our customers smoothly integrate this next generation of high-performance technology as we move into more connectivity and more challenges into protecting information against upcoming technological threats.

*References:*

- C. Abellan and V. Pruneri, “The Future of Cybersecurity Is the Quantum Random Number Generator", IEEE Spectrum (2018)
- C. Abellan et al., "Quantum entropy source on an InP photonic integrated circuit", Optica (3) 9 (2016)
- C. Abellan et al., "Fresh and pure quantum random number generation for loophole-free Bell tests", Phys. Rev. Lett. 115 (2015)
- C. Abellan et al., "Ultrafast quantum random number generation based on the accelerated phase diffusion process in semiconductor lasers", Opt. Express 22 (2014)
- M. Mitchell et al., "Strong guarantees in ultrafast quantum random number generation"; Phys. Rev. A. (2014)
- B. Hensen et al., “Loophole-free Bell inequality violation using electron spins separated by 1.3 kilometers”, Nature 526, 682-686 (2015)
- M. Giustina et al., “Significant-loophole-free test of Bell’s theorem with entangled photons”, Phys. Rev. Lett. 115, 250401 (2015)
- K. Shalm et al., “Strong loophole free test of local realism”, Phys. Rev. Lett. 115, 250402 (2015)
- M. Rude et al., “Interferometric photodetection in Silicon photonics for phase-diffusion quantum entropy sources”, Opt. Express 26 (24), pp. 31957-31964 (2018)