The two sides to quantum physics

Quantum physics is particularly important in regard to the security of secret information in two respects. First, it is possible that it will be relatively easy in future to crack today’s encryption systems with quantum computers. In the past week, citing documents from Edward Snowden, the Washington Post reported that the NSA is pursuing a research programme to build quantum computers. It is, however, clear to experts that this is a long-term goal that will not be implemented for at least another 20 years. Second, quantum physics and the field of quantum cryptography are creating new possibilities for the secure encryption of data, and initial applications already exist in this area. This is also of interest to the NSA, as Renato Renner, Professor at the Institute for Theoretical Physics, says in an interview with ETH News.

ETH News: The American intelligence services are obviously interested in quantum computing – why?
Renato Renner: Quantum computing is a promising area of research. If a quantum computer exists someday – and we are still at least 20 years away from this – then it will be able to solve certain mathematical problems far more quickly than standard computers. Today, information is often encrypted using public-key encryption. This system is used not only by the state, but also by private organisations to send emails securely and for e-banking. Intercepting data encrypted in this way calls for a computing effort that cannot be managed in a reasonable amount of time by standard computers. The problem is one of those mathematical tasks that a future quantum computer would be able to solve very quickly. It is therefore not surprising that this of interest to the intelligence services.

Quantum physics, however, not only promises a revolution in terms of decryption but also in the encryption of information – with the key word being quantum cryptography...
Yes and, interestingly, public discussion is currently paying little attention to this field, even though market-ready applications already exist here, unlike in quantum computing.

What does quantum cryptography achieve?
It enables a fundamentally different encryption system that is not based on the difficulty of mathematical problems. Instead, quantum physics is used to ensure that one cannot be spied on unnoticed, even with a future quantum computer. The NSA is also interested in quantum cryptography; this technology is mentioned several times in the documents from the authorities that were published on the internet by the Washington Post. However, I think it will be more of a curse than a blessing for spies: the anti-interception security offered by quantum cryptography systems can be mathematically proven based on the laws of quantum physics. And even the fastest computer in the world cannot elude the laws of physics.

What are the main differences between quantum encryption and today’s standard methods?
In the case of the public-key encryption widely used today, keys exist as classical digital information and the system operates with two keys, a secret key and a public key. Several people – anyone in possession of the public key – can, for example, encrypt messages, while only the owner of the secret key can decrypt it. As the secret and public keys are connected, however, it is theoretically possible to calculate the secret key from the public one. But this calculation is very tedious using today’s computers. In quantum cryptography symmetrical keys are used; this means that the sender and recipient of the message use the same key, which therefore must be kept secret. The challenge is to generate this type of key and distribute it to the sender and recipient without it being read by anyone else.

How is this achieved with quantum cryptography?
The key is coded in the polarisation of light particles and then transferred from the sender to the recipient via optical fibre. Due to the quantum physical nature of the light particles, any attempt to measure their polarisation during the transmission would lead to a change that would be noticed by the recipient.

Does this mean that anyone interested in information security as a practitioner today should focus on quantum cryptography rather than quantum computing, as the latter is still a long way off?
Even if quantum computers are likely to be built only in 20 or 30 years’ time, quantum computing has an extremely practical relevance today. If you want to encrypt data now that must still be protected against unauthorised access in 2035, then you need to use encryption processes that cannot be cracked by a quantum computer. Intensive research is already being carried out into such alternative processes and quantum cryptography is one of these.

You have said that market-ready quantum cryptography applications already exist. What are these?
In reality, quantum cryptography is one of the first commercial applications of quantum technology. ID Quantique, a spin-off from the University of Geneva, is the first company to market a functional quantum cryptography system. Its system consists of two terminal devices – one for each of the partners who want to communicate in a secure way – that must be connected via a direct fibre optic link. ID Quantique is one of the leading companies in this field worldwide; its technology is based on theoretical principles developed at ETH Zurich.

What kind of principles?
The anti-interception security of quantum cryptographic devices cannot be shown in principle using experimental tests. Instead, mathematical proof is required to show that no offensive strategy can be used to intercept the transmitted keys silently. We developed the mathematical principles for such a security proof in my research group.

Can this system already be used?
Yes. ID Quantique’s list of clients not only includes universities that use the system for research purposes but also banks. However, in practice, the transmission distance is currently limited to about 50 to 100 kilometres because there is still no way of amplifying quantum physical information. But research is being carried out here and, in any case, there are fewer technical obstacles in quantum cryptography than in quantum computing.