Chasing the holy grail of “un-hackable data” with quantum key distribution

by

16 February 2017
Alex Ling

As quantum computing transitions from theoretical physics into actual devices, the ability to encode information and compute at atomic or sub-atomic levels is presenting new challenges, and potentially new solutions, for encryption.

Encryption is the encoding of messages so that only authorised parties with a valid cryptographic key are able to access them.

One of the issues with encryption lies with the sharing of keys. The physical exchange of keys, while secure, is a cumbersome process because it involves the parties having to meet up.

The alternative has been to use a public key system whereby a public key is used to encrypt the message, which can then be deciphered only by a party with a private key that is known only to him.

Speaking at EmTechAsia 2017, Dr Alexander Ling, principal investigator, Quantum Optics, at the Centre for Quantum Technologies, National University of Singapore, pointed out that the public key infrastructure relies on mathematical complexity to make it difficult for hackers to extract the private key. However, this approach is based on the difficulty of computation, and quantum computing could transform the equation totally.

Quantum computing unleashes processing power by many orders of magnitude. Unlike traditional computing where data is stored in binary digits and processing power is limited by the number of binary transistors that can be packed into a system, quantum computation makes use of quantum bits (qubits) which exist not just as a 0 or a 1, but also a 0 and 1. This means that a qubit will be able to perform two equations at the same time, two qubits can perform four, three qubits can perform eight, and so on.

As control over qubits improves and techniques evolve to store and transmit them in large numbers, the stage is set for the development of the quantum computer, which Ling described as “a specialised piece of hardware built to solve special problems”.

One of these problems, incidentally, is prime number factorisation, which is the basis of the public key infrastructure. What this means is that quantum computing could up-end the security of encryption systems that rely on the public key infrastructure.

At the same time, however, parallel developments in the quantum computing space could hold the key to solving this issue.

In his presentation, Dr Ling discussed the possibility of “Un-hackable Data with Quantum Key Distribution”. The core technology behind the proposed solution is a quantum entanglement generator which splits a ray of light into correlated photons to form the encryption key. If one photon has a polarisation pointing in a particular direction, the twin photon will also point in the same direction. If, however, there is an eavesdropper, the correlation between the proton pairs deteriorates. The error rate can thus be calculated to assess if it exceeds an acceptable threshold, in order to determine if the key is secure.

There are many advantages to this approach of distributing the encryption key, said Ling. As photons or single particles of light, quantum particles can be transmitted through optical fibre, addressing the issue of key distribution. For example, data centres can be secured using quantum keys, which are then exchanged over an optic fibre connection.

“As long as Alice and Bob share a link, we can automate the process of key distribution. They do not have to physically meet.”

According to Ling, the proposed solution is also immune to computing attacks. “Quantum signals are generated based on intrinsically random physical processes, which no amount of computing power can predict,” he said. 


CONVERGENCEASIA PIONEERS SERIES 

Prof Gordon Bell on "exascale nirvana" and other supercomputing trends

Gordon Bell of supercomputing's Gordon Bell Prize talks to ConvergenceAsia about the "scaling up" and "scaling out" of HPC and its new applications in analytics and data science. Bell was in Singapore recently for Supercomputing Frontiers 2017, where he delivered a keynote on "Three Decades: Motivating and Measuring HPC Progress". 

  • Video by Chan Wai Peng