Quantum mechanics and Blockchain to secure IoT

Question

I came across this article on IoT Security with Blockchain and Quantum mechanics.

So, what's a quantum-secure blockchain and why is it more secure in an IoT setting?

Answer 1

Quantum mechanics and the blockchain

Those are two evergreens in every futuristic buzzword bingo. Thus, naturally they get mixed with the Internet of Things, which has been derisively put in the same category of futuristic stuff people half get.

So, let's sort the stuff out a bit. I'm assuming that you're referring to this article Aurora dug up. Unfortunately, the site throws all IoT technologies into one box. Yes, blockchain can be considered an IoT technology and there are many people right now thinking about employing blockchains in IoT. (E.g. Yet another blockchain IoT consortium)

Following up through the cited news article we land at a whitepaper about "Quantum-secured blockchain[s]." Obviously we need some basics to understand that whitepaper, which not unusual for whitepapers assumes quite a lot of pre-existing knowledge about quantum mechanics, quantum computers, algorithms running on quantum computers, encryption methods, their weaknesses when attacked by quantum computers, quantum key distribution (QKD)—and last but not least, blockchain.

This list makes it impossible to go into every detail of the problematic and most of what I listed has Wikipedia pages that are quite detailed. We need just the following information:

• Quantum mechanics are really weird. Over at Physics.SE there are 12k questions about those things, because, as I said weird. The interesting weird property for us is:

  This results from a fundamental aspect of quantum mechanics: the process of measuring a quantum system in general disturbs the system. (QKD Wikipedia page)

• Quantum Computers (which don't exist yet) can use different algorithms then regular CPUs. There are algorithms like Shor's algorithm which can make short work of classical encryption.
• Current cryptography is basically just posing mathematical problems that are to hard to solve using conventional computing. More to the point, the time needed to crack the code is sufficiently long to reach a point in time where the information is valuable to an attacker.
• Blockchains use—very simply put—classic encryption methods for their security.

What does that mean?

We could stop here. There are no quantum computers. Our current cryptographic methods and standards are quite fine. (Consider Skeptics.SE if you're wearing a tin foil hat due to any existing quantum computers in certain three-lettered organizations' basements.)

But if there were quantum computers?

If we had to protect our IoT applications against a quantum computation enabled attacker the technology is either relatively common or you have really high security standards. Either way, for the sake of our quantum-attacker-proof IoT blockchain it doesn't matter, we want it secure.

What's the solution of the whitepaper offering us and how does it apply to IoT?

After a few discarded approaches the core is this:

In the present work, we describe a blockchain platform that is based on QKD and implement an experiment demonstrating its capability in a three-node urban QKD network.

(Page 2)

The whitepaper describes a blockchain platform that uses quantum key distribution. Great. Thus, they do not encrypt any transmission of anything in the blockchain. They "just" switch the authentication key for the signature.

Waaaait, what? You're saying? Why would a block chain care about that? Authentication is irrelevant in blockchain! That's correct the participants in classical blockchain computation are not authenticated. In fact many argue that blockchains that aren't open and permissionless aren't really block chains. Even the paper says:

The utility of QKD for blockchains may appear counterintuitive, as QKD networks rely on trust among nodes, whereas the earmark of many blockchains is the absence of such trust.

(Page 2, emphasis mine)

Of course, there are people who say, every chain of blocks is a blockchain.

The blockchain is a distributed database in which the records are organized in a form of consecutive blocks.

(Page 4, emphasis not mine)

There is another major drawback which can be derived from these two excerpts:

In summary, we have developed a blockchain protocol with information-theoretically secure authentication based on a network in which each pair of nodes is connected by a QKD link.

(Page 3)

The basis for our experimental work is our recently developed modular QKD device [25, 35–38] driven by a National Instruments NI PCIe-7811R card. This setup uses a semiconductor laser LDI-DFB2.5G controlled by an FPGA board Spartan-6 to generate optical pulses at the standard telecommunication wavelength 1.55 m and a 10 MHz repetition rate. We have used ID230 single-photon detectors from ID Quantique.

(Page 4)

Short, every node has an OKD link. Thus, every node has a photon detector. That's also were this get's really un-IoT-y. Each IoT device to employ this kind of blockchain would need a photon state sensor (as photons are the most common quantum state transportation medium).

Summary

Current blockchain implementations have a level of separation between the edges (i.e. the guy who buys stuff in bitcoins on a device) and the computational entity creating new blocks (i.e. miners).

The authenticity of any transaction in a classical blockchain is derived by miners (via proof-of-work). The transaction itself can be easily created by edge devices.

Employing QKD allows only participants with valid quantum keys to add transactions. That means highly tuned quantum state sensory equipment at each transaction capable device.

TL;DR—Quantum what?!

Interesting research project, but there are currently no potential attackers and even if there were the hardware would make it economically unfeasible for the Internet of Things.