Random Number Generator

About

Encryption is vital to protect sensitive data by converting it so that cannot be read by humans or computers without access to secret numerical keys. Kerckhoffs’s principle states that the security of a cypher must reside entirely in the key – i.e. it must be entirely random. High quality random numbers are therefore the essential foundation and enabler of encrypted-communication systems and products. However, high-quality random numbers are hard to produce so, currently, the vast majority of cryptographic systems use algorithmically based random number generators (RNG), the so called “Pseudo-RNG” (PRNG). The weakness of this approach is that PRNGs are entirely deterministic: an eavesdropper who obtains the initialisation settings of the algorithm will be able to predict all possible outcomes, thus compromising the entire protocol. This vulnerability makes PRNGs one of the principal targets of hacking attacks. In a true random number generator (TRNG) randomness comes from a naturally occurring source of randomness, i.e. entropy. Because the outcome of quantum-mechanical events are intrinsically random, they are considered the ‘gold standard’ for random number generation (QRNG). QRNGs have typically been based on specialised hardware, such as single-photon sources and detectors, resulting in systems being cumbersome, large and high-end/expensive. Crypta Labs QRNG is based on the quantum properties of light. The figure below sets out the principal of our technology: a light source, such as an LED, emits a controlled and attenuated flux of photons, which is directed on a sensor of a digital camera. The camera takes a picture and stores a frame in an uncompressed file format. This matrix of raw numerical values corresponds to the number of photons which were detected during the exposure time of the picture. The raw data is then processed by our algorithm to enhance the uniformity and independence of the numbers ready for use in cryptographic applications. Crypta Labs have demonstrated that sensors for image acquisition of mobile devices have reached such a level of sensitivity that they can differentiate the number of photons hitting the sensor to the single photon level. We have also recently proven that the image taken by the camera can be converted into a bit stream and that the raw data is driven by the quantum effects of light and not technical, non-quantum noise.