Random numbers are the key to strong digital security. The more unpredictable the numbers generated, the tighter the encryption and thus the more secure the transaction.
A research team at Quantum Base, a company spun out of Lancaster University in the U.K., has created a “small, low-power device that produces pure random numbers,” Quantum Base’s CEO Phillip Speed said in earlier this month.
Why is that important?
Because the blockchain universe, primarily the technology underpinning blockchain and digital currency wallets, could benefit from stronger security thanks to those random numbers. If you’re a bitcoin miner, don’t you want your transactions to be as secure as possible?
Here’s the science: Pseudo Random Number Generators, or PRNGs, tend to be software solutions, such as bitcoin’s cryptographic algorithm Elliptic Curve Digital Signature Algorithm (ECDSA). These systems have inherent flaws that expose them to predictability. Hardware-based RNGs are arguably a superior alternative, but both options rely on rules and inputs subject to influence from external factors. Quantum Random Number Generators (QRNGs), which are hardware devices that rely on the theory that subatomic particles are intrinsically random in behavior, tend to be slow and expensive.
According to in the journal Nature that lists two Quantum Base executives among its co-authors, the company’s invention appears to have three notable characteristics that potentially render QRNG a viable alternative for wide-spread implementation instead of hardware-based RNGs: smaller size, faster speed and lower cost. These factors allow for easy integration of the device into microelectronics.
Harnessing Subatomic Randomness
Split a beam of light in two by shining it through a piece of glass so that part of the light is reflected and part is transmitted through the glass. The theory of quantum mechanics says that as the photons hit that glass, there is no way you can predict which way they will go. You could, however, with the right tools, observe the path traveled by those photons.
You could then exploit this intrinsic uncertainty of the photons by randomly generating 1s and 0s, the foundation of computer logic, based on which of the two outcomes ultimately happen. Researchers have been working for almost two decades to capture this sort of natural uncertainty to achieve true random number generation.
Swiss company has been developing QRNGs since 2001, and a QRNG sponsored by generates random numbers in real time and is publicly accessible via a web server. In general, the practicality of previous QRNGs in wider cryptography implementation has been limited by factors such as device size and cost, although ID Quantique now markets called Quantis.
Robert Young of Quantum Base tells ThirtyK his company’s device overcomes the traditional challenges by extracting random numbers from quantum tunneling using a simple semiconductor structure, called a resonant tunneling diode (RTD). In quantum tunneling, a particle tunnels through a barrier that it would not be able to overcome according to classical physics.
The diode, a thousandth of the width of a human hair, use a current instead of light and could be incorporated into both new and existing microelectronics. Ultimately, the output can be directly used as a random stream of bits or can be further distilled using randomness extraction algorithms.
In contrast, other hardware based RNGs, like the one you might have on your smartphone, are typically “set up to work external to the main processor, usually a separate chip which sits alongside the processor,” Young says. But this separation increases cost and power requirements. Moreover, he adds, it “compromises security with slower speeds of many orders of magnitude lower than the clock rate” and opens the door for man-in-the-middle attacks, in which the random numbers are measured or influenced as the beam communicates between two chips.
Next Stop: Commercial Testing
Helmut Katzgraber, a quantum computing professor at Texas A&M University computing professor and a principal researcher at Microsoft Research, tells ThirtyK that “the potential of the new approach to be fast is what sets it apart.” Excitement should be reserved, however, until there is “proof that the new QRNG will pass the [U.S. National Institute of Standards and Technology] tests once it is run at GHz speeds instead of kHz,” he adds.
Quantum Base acknowledges in the paper that tests at higher speeds are definitely in order and the electronics required for the commercial testing have to be developed. Young says he believes this benchmark will be achieved. The diodes Quantum Base used in the research “operate very quickly … much faster than modern processors,” he says.
Even if Quantum Base’s invention stands up to the scrutiny of NIST, there might still be skepticism that the QRNP is bulletproof. “There could be subtle correlations that might only show up when it is used for actual applications,” says Katzgraber. He recommends the device “first be implemented at GHz speeds across several applications before it can be deemed reliable.”
Successful and practical QRNG could provide, at least theoretically, unbreakable quantum security. There could be benefits to the blockchain universe primarily to the cryptography underpinning blockchain and digital currency wallets.
Bitcoin mining is known for consuming time and . Young says that fully integrating a RNG device into system hardware could create universally trusted random number generation. Then, some time in the future, he adds, we might be able to look forward to “the replacement of the process with proof of randomness.”