Perfect secrecy cryptography via mixing of chaotic waves in irreversible time-varying silicon chips

erek

erek

[H]F Junkie
Joined
Dec 19, 2005
Messages
10,951
Scared of Quantum computing cracking all of classical encryption algorithms?

"Protecting confidential data is a major worldwide challenge. Classical cryptography is fast and scalable, but is broken by quantum algorithms. Quantum cryptography is unclonable, but requires quantum installations that are more expensive, slower, and less scalable than classical optical networks. Here we show a perfect secrecy cryptography in classical optical channels. The system exploits correlated chaotic wavepackets, which are mixed in inexpensive and CMOS compatible silicon chips. The chips can generate 0.1 Tbit of different keys for every mm of length of the input channel, and require the transmission of an amount of data that can be as small as 1/1000 of the message’s length. We discuss the security of this protocol for an attacker with unlimited technological power, and who can access the system copying any of its part, including the chips. The second law of thermodynamics and the exponential sensitivity of chaos unconditionally protect this scheme against any possible attack."

https://www.nature.com/articles/s41467-019-13740-y
 
all 53 qubits

upload_2019-12-22_1-13-13.jpeg
 
  • Like
Reactions: erek
like this
Mega6 said:
all 53 qubits

View attachment 210145
Click to expand...


* Now talking on ##security
* Topic for ##security is: Welcome to Computer, Network & Physical ##Security | Please stay on topic
<erek> did you guys review this already @ https://www.nature.com/articles/s41467-019-13740-y ?
<erek> anyone?? https://www.nature.com/articles/s41467-019-13740-y
<spinull> nobody reviewed something that was published in nature 2 days ago!?!? I just cant believe it
<spinull> what do you guys even do all day?
<Zalyssa> anyone??
<plasticuproject> I'm skimming it now. But normally I just browse instagram influencers all day.
<shadowdaemon> lol
<shadowdaemon> I'm no crypto expert, but it seems a bit too good to be true.
<shadowdaemon> "Beyond the initial communication required for authenticating the users, the system does not require electronic databases, private keys, or confidential communications."
<erek> :)
<shadowdaemon> So how does the initial exchange take place?
<spinull> obviously it uses dark energy to transmit on the quantum level
 
<spinull> obviously it uses dark energy to transmit on the quantum level

Classic.
 
"Perfect secrecy cryptography via mixing of chaotic waves in irreversible time-varying silicon chips"

I'll betcha U can't say that 3 times (fast) :rolleyes:
 
<Raito_Bezarius> erek: the first paper works under strong assumptions of thermodynamics
<Raito_Bezarius> At the quantum level, it's not obvious those are still true
<Raito_Bezarius> But at the same time, the paper uses some bits of quantum theory
<Raito_Bezarius> It is strange
<Raito_Bezarius> You have scale issues because quantum theory and thermodynamics are disconnected
<Raito_Bezarius> Let's admit that their physics is sound because this paper is mostly on physical infeasibility and not mathematical infeasibility which is weaker than the latter
<Raito_Bezarius> It looks like that it'll require specific environments and conditions for it to work like they expect it

<Raito_Bezarius> I suspect there could be a catch which could open up to a quantum attack if you were an attacker with unlimited resources, like: you cannot replicate the state but you can get the most probable one, given enough samples, the method will recover up to X % of the bits, and after that, you can perform the full recovery in certain cases
<Raito_Bezarius> It's cute but if I controlled physics, I could break it ; on the other hand, discrete logarithm problem don't really depend on the physics parameters of the world, even if I changed the speed of light, I cannot break it (except if I know a mathematical solution)
 
Quantum computers cant even factor 15 with better than 50% reliability. I've become very skeptical that they will ever be practical for anything.
 
I think the issue with annealing factors might be iteration?
Would have to unroll a bidirectional multiplier flat. Quickly
takes an unreasonable number of qubits. If you try to re-
use any of those pieces, the input can't feel the output.
Probably cause they now happen at different times.

I once thought about trying that with LC room temerature
audio frequency parametrons with some kinda automatic
gain control that holds each in a simulated superposition
just below Gain=1, till it was time to freeze out an answer.

My theory was: If it solves something impossible, all day
long to get an answer might actually be fast? But unrolled
anything takes way too many parametrons. Also have to
consider phase shift across a large machine. I'm thinking
far less than a quater wave there and back, or it can't sim
quantum annealing...

Even if you get two random factors, a lot of times its going
to be 1 and N. Which is useless unless there are no other
factors (prime). But how many times must you try and fail
before a true prime (not just bad luck) can be proved?
 
Last edited:
  • Like
Reactions: erek
like this
Wat said:
If entropy is always increasing, where's it coming from???
Click to expand...
The level of disorder in the universe is steadily increasing. Systems tend to move from ordered behavior to more random behavior.
 
Mega6 said:
The level of disorder in the universe is steadily increasing. Systems tend to move from ordered behavior to more random behavior.
Click to expand...

that's why i'm hoping to totally virtualize the clock signal generator in favor of the obsolete quartz crystal and other hardware based clocks
 
Or just spray some Super 77 that Bendix drum and wrap it in VHS tape...
Maybe in a spiral, so the ends are at the edges where a dab of epoxy
to keep it all from unravelling seems less likely to upset the heads.
New surface with one predictable discontinuity per track.

Or some vintage tape from that machine beside it? I've tried and failed
to roll memory cores from VHS. Scotch taped into straws, then cut into
small donuts. Measured no hysteresis whatsoever. Still no clue where
it went wrong. Any old magnetic tape is not always that simple...

VHS was made to be read/written on a slant while wrapped on a drum
at some angle. Perhaps in the drum case it might still be plausible?
 
Last edited:
erek said:
that's why i'm hoping to totally virtualize the clock signal generator in favor of the obsolete quartz crystal and other hardware based clocks
Click to expand...
Environmental conditions are more likely to change your hardware quartz (or other) based signal
 


Though I warn, you still need to lock onto a reference thats stable over long time period.
This just eliminates worry of how that reference misbehaves over the short time period.
Its a progrmmable frequency synthesizing VCO with vastly improved jitter. Not quite a
standalone replacement for quartz unless you don't care how it drifts slowly over time.

An add-on that makes quartz better still wants for a temperature compensated or oven
stabilized crystal. If you really want to kick quartz to the curb, go Rubidium or Caesium.
In which case, you still need a jitter free VCO and PLL to synthesize useful frequencies.
https://en.wikipedia.org/wiki/Rubidium_standard
https://en.wikipedia.org/wiki/Caesium_standard
 
Last edited:
Mega6 said:
The level of disorder in the universe is steadily increasing. Systems tend to move from ordered behavior to more random behavior.
Click to expand...

That relies on a rather arbitrary definition of 'order' though
 
KD5ZXG said:


Though I warn, you still need to lock onto a reference thats stable over long time period.
This just eliminates worry of how that reference misbehaves over the short time period.
Its a progrmmable frequency synthesizing VCO with vastly improved jitter. Not quite a
standalone replacement for quartz unless you don't care how it drifts slowly over time.

An add-on that makes quartz better still wants for a temperature compensated or oven
stabilized crystal. If you really want to kick quartz to the curb, go Rubidium or Caesium.
In which case, you still need a jitter free VCO and PLL to synthesize useful frequencies.
https://en.wikipedia.org/wiki/Rubidium_standard
https://en.wikipedia.org/wiki/Caesium_standard
Click to expand...


i want to digitally record the hardware clock generator signal and use only a virtual representation of the hardware only
 
KD5ZXG said:
I think the issue with annealing factors might be iteration?
Would have to unroll a bidirectional multiplier flat. Quickly
takes an unreasonable number of qubits. If you try to re-
use any of those pieces, the input can't feel the output.
Probably cause they now happen at different times.

I once thought about trying that with LC room temerature
audio frequency parametrons with some kinda automatic
gain control that holds each in a simulated superposition
just below Gain=1, till it was time to freeze out an answer.

My theory was: If it solves something impossible, all day
long to get an answer might actually be fast? But unrolled
anything takes way too many parametrons. Also have to
consider phase shift across a large machine. I'm thinking
far less than a quater wave there and back, or it can't sim
quantum annealing...

Even if you get two random factors, a lot of times its going
to be 1 and N. Which is useless unless there are no other
factors (prime). But how many times must you try and fail
before a true prime (not just bad luck) can be proved?
Click to expand...

Did anyone else read this and think it was one of those "keep pressing the middle autocomplete suggestion" memes?
 
You must log in or register to reply here.
Back
Top