What role does quantum computing have in defence, any, or none?
The huge advances in supercomputing over the last few years have benefitted a number of industrial, financial and scientific communities, but there has not been a great deal of media about them being used in defence and security applications.
Cryptography, aerodynamics, social media connection analysis and other large data set correlation are, of course, the most obvious answers, but are there any others, others that involve really hard sums!
Watch this video from D-Wave Systems whilst you are thinking about it.
15 Responses
Interesting. OK here’s a few ideas:
Pattern analysis in urban insurgency (interpreting Gorgon Stare and fusing with e.g. Sigint data)
Optimising ingress/egress routes for low-observable aircraft vs complex IADS
Theatre weather forecasting
Modelling littoral radar scattering
Modelling thermal channels etc in water for advanced ASW (or to enable subs to hide better)
Fusing ambient radio scattering for passive stealth detection
Identifying unusual traffic/computer usage patterns that may indicate cyberattack
Solving the problem of procurement overspend (just kidding)
Really cool! No – really – I didn’t know we had the capability to get so close to absolute zero. That’s if the number in the video (0.015K) is for real.
I remember reading about Josephson Gates back in the 80s, when they were discounted based on the need for a building-sized cryo plant needed to cool the 4 inch cube supercomputer. Good to see the technology moving forward, although its unlikely to ever break into the industrial or consumer market. (Risky making technology predictions – the boss of IBM is on record as having predicted the global demand for computers may have been as many as five.)
I think the number one area that will be impacted by quantum computing will be SIGNIT. Quantum computing makes current data encryption useless. It’s no secret that the NSA and GCHQ have been pioneering quantum computers. Beyond that its likely to affect every military system in the way the microprocessor did. Smarter missiles, aircraft and autonomous systems.
It may also have implications for applications like active stealth where vast amounts of computing power may be required at Super fast speeds.
Martin – I see no mobile applications of superconductor quantum machines unless there is a huge step forward in cooling systems. All fine in the basement of security centres, possibly might even make some of the bigger corporations for explosion simulation, CFD, hypersonic flight dynamics. But embedded on one-shot projectiles? Highly unlikely in my view.
@DOMS
“Solving the problem of procurement overspend (just kidding)”
Ain’t no ‘puter that powerful
Having been interested (at an informed layman’s level) for some time, in particle physics, the claims implied by ‘D-Wave’ have me puzzled. The latest data from sources various give a hopeful time of fifteen years before a workable computer, based on q -bits for processing could become reality. Even keeping q-bits in a quantum state is in itself a leading edge achievement. Reading the output from various particle physics study groups show a fair helping of scepticism regarding a fully quantised computer. Of course quantum effects take place in most electronic gadgets, in fact they wouldn’t work without them, however prefixing everything with the word ‘quantum’ does not make them so.
While Quantum computing may be one or two decades away, once we get to units sized to fit in a field HQ, I can see real time modelling of projected enemy routes of advance, and accurate positioning of optimum defensive points and choke points a reality down to Battn. level. It could revolutionize battle tactics, as defensive positions could be modeled for effectiveness almost instantly.
Quantum computing don’t exist, and stands at the point that it has not been conclusively proven impossible.
D-Waves machines are not quantum computers, they are computers optimised to run certain quantum calculations faster than general purpose computers. Some of the massive IT players have picked up a few to see what this means in a data analysis enviroment, (and to play with, just because it’s interesting) but they haven’t ordered anything more than a handful of machines to keep the company going.
You’ll soon know when Quantum computing is considered possible: All hell will break loose within the IT industry. It’s possible to secure against quantum computing breaking encryption with technologies now considered obselete, but the chaoes switching will be interesting to watch. I’m hoping i’ll have retired by then.
@Dangerous Dave “While Quantum computing may be one or two decades away, once we get to units sized to fit in a field HQ, I can see real time modelling of projected enemy routes of advance, and accurate positioning of optimum defensive points and choke points a reality down to Battn. level. It could revolutionize battle tactics, as defensive positions could be modeled for effectiveness almost instantly.”
I don’t think you need anything like quantum computing to do that, in fact I’m quite surprised to hear that it’s not happening already. I speak from some experience.
I’m an IT guy and back in 1985/1986 I was involved in a project originating from research at DRA Malvern & subsequently resulting in what was for me a huge piece of business from the RAF. The basic concept wasn’t classified although the computer program itself and the data in it most certainly were.
The basic concept was a battlefield-deployable (and subsequently deployed on a world-wide basis) very powerful (at the time) workstation computer that would be sent to field command locations. The program had accurate contour maps of wherever the unit was deployed and also mathematical models for the attack envelopes of the then-current portable surface-to-air missiles (Rapier I think). The idea was that the commanders could highlight on the maps the key areas they wanted defending, tell the system how many defensive units they had available, and the computer program would tell them the optimal placings for the units to get the best coverage considering the local terrain. No more details were declared to me at the time so I don’t know (or have forgotten) whether the nature of the threat (e.g. enemy aircraft type) was also taken into consideration and whether it also attempted to optimally site other defensive assets (e.g. AA guns). One thing I do remember however, because it stuck in my mind as an illustration of the realities of war, was that the system would consider two attack envelopes for the Rapiers, one if they were fired in “safe” mode with a view to avoiding any collateral damage to any friendly forces near the launcher and the second in full war mode where the safety aspects were ignored and the missiles allowed to be launched on a far wider range of inclinations and the consequences of who might get caught in the back-wash ignored.
Given the above was mid 1980s technology, although admittedly cutting edge, I would think that one heck of a lot of position modelling could be done in near realtime with a relatively cheap off-the-shelf (but obviously militarised) hardware. A small server rack of today’s technology, about the size of a 2-drawer filing cabinet, would probably have something like 25,000 times the computing power of the workstations we were using for that RAF project.
Re quantum computing per-se, and speaking both as a computer scientist and a physicist, I would say that (a) it’s still going to be a while before it’s a reality, and a while longer before cooling issues make it practical for it to leave machine rooms, (b) there is still controversy about whether the D-Wave really is a true quantum computer so arguably they don’t exist at all yet, and (c) if and when they do come along the impact on encryption, namely the ability to break it with great ease, will be immediate and dramatic.
– Julian
You TD commentators belong to the smart party, am impressed!
This sounds a bit less complicated than a RF network planning package, and those will run on a laptop.
@Alex – “This sounds a bit less complicated than a RF network planning package, and those will run on a laptop.”
Indeed, but don’t forget that I was talking about 30 year old technology.
With the ability today to deploy portable hardware that is tens of thousands of times more powerful, in fact if one factors in GPU computation and the application is able to use it we’re probably talking about filing-cabinet sized hardware that is hundreds of thousands of times more powerful than the 30-year-old workstation I was referring to, we are probably already limited by factors other than computing power in terms of doing in-field stuff. One “factor other than computing power” that occurs to me is the availability of input data. Even in the case of my 30-year-old example the system was pretty useless if suitably detailed maps of the deployment area weren’t available.
Garbage in garbage out comes to mind. Tactical options need tactical thinking and while there are a lot of variables the problem is that some are not quantifiable. Not forgetting that you don’t control what the enemy will do, as the old saying goes ‘No plan survives contact with the enemy’. If the enemy fails to conform to the assumptions you put into your computer then you are heading to a problem. Good planning is nuanced and takes this into account, and is dependent on military experience, the activity of planning also helps spot tangential gotchas and enables assumptions to be challenged.
I suspect the GBAD modelling at that time was primarily concerned with the debate over more VSRAD missiles or guns. Determining coverage by computer is useful but not reliable due to things like trees growing, maps not being fully up to date, etc. In the end their is not substitute for a competent recce offr and boots on the ground.
The timeframes to quantum computing happening just got cut by two thirds yesterday, from what most comments were indicating: http://www.nature.com/nnano/journal/vaop/ncurrent/full/nnano.2015.261.html
to 4-5 years from now.
Will be very practical to field, though (outside snooping centres and their unlit machine halls):
” a promising result especially when combined with the fact that the device was built in silicon, fabricated with standard industry processes that lend themselves to miniaturization and scaling up the number of qubits. (Like similar quantum computers, this device still needs to operate at temperatures near absolute zero and under a strong magnetic field; but since quantum computers are not meant to replace traditional computers altogether, this should not be a big issue.)”
As I previously inferred, the hard part is in maintaining the quantum state of the cubit in a manner that would lend itself to any hint of practical use – especially in combat conditions. When that is achieved, tremendous possibilities can at last be realised, especially in the medical and pharmaceutical fields.
Comments are closed.