encryption and cryptography in Trek canon

[jefferiestubes8]

What types of encryption has there been used in canon?
In STII:TWOK we had Kirk do a biometric iris scan. I'd like to discuss the encryption itself and not biometric or password authentication.

And a larger discussion where will cryptography be in 50, 100, 300 years?
We currently have AES256 bit encryption with 4096-bit encryption keys commercially available now in 2009.
And we all know the strength of a password is very important. (alpha, numeric, symbols mixed, upper and lower cased, and no dictionary words)
I'd like to discuss the actual encryption itself though.
If it would take many years now to break a 128-bit key what about the computers in the future and the needs of cryptographic keys?

 

[Timo]

The iris scan may have been but part of a wider-scope recognition protocol; at the very least, the computer would also have analyzed Kirk's audio print, and it may also have required the presence of some sort of a key device on Kirk's person. But the actual encryption is something that remains hidden from the audience...

Considering that future computers are extremely good at deciphering alien languages at breakneck speed, one might be tempted to think that all encryption systems based on a fictional language (in essence, all the encryption systems so far used by mankind) would be way too easy to break. Some sort of a conceptual change would have to be introduced: self-protecting codes, perhaps, active software entities that will adapt to breaking attempts and will secure themselves against the attack. Or codes that aren't mere abstractions, but make use of physical elements (similar to the quantum-entanglement code concepts being thrown around nowadays).

That said, Spock's father was credited with inventing or otherwise devising what was called a "satchel code" in the novel Spock's World. This supposedly was a practicable means of encryption at the time, not merely a meaningless hobby project for the Ambassador. Make of that whatever you will; for all we know it might refer to an active defensive software around the message, or to a physical package around the abstract message content.

Timo Saloniemi

 

[Myasishchev]

Well, in real life, we're seeing the experimental technology of quantum computers coming closer to fruition. A functional quantum computer would make basically all known encryption obsolete, since the strong forms of encryption currently in use rely on the intractability of certain mathematical problems, namely the difficulty in factoring large numbers--the 128-bit key mentioned would take a classical computer longer than the age of the universe to factor. However, the problems are no longer intractable when the computation is shared between a large number of universes (or for Copenhagen interpretation fans, however you want to see it), and the prime factors used as keys will be busted wide open.

Now this doesn't spell the end of privacy entirely. Another interesting quantum phenomenon is that observation affects the information carried by any particle or set of particles. Listening to a message detectably changes the message. Thus, while private communication may not be possible, eavesdropping undetected is even less feasible. An eavesdropper determined to hear the message could only enforce silence against a speaker determined to keep his message secret.

Presumably Star Trek computers are quantum and not classical.

 

[Timo]

...Of course, in Star Trek there may exist technologies to defeat the observer effect. Devices like transporters or cloaks might depend on such technologies, but a spinoff would be a device that can defeat the quantum-state encoding.

Timo Saloniemi

 

[Myasishchev]

^Transporters preserving quantum state information is totally unnecessary anyway, insofar as the quantum state of any beamed object is going to change in the moment following beaming anyway.

Cloaks would benefit from it, but cloaks are shown to be vulnerable to exactly that sort of thing, with fields of whatever-particles strung between two ships capable of identifying the predictable changes the cloak imposes on those particles.

 

[noknowes]

The iris scan may have been but part of a wider-scope recognition protocol; at the very least, the computer would also have analyzed Kirk's audio print, and it may also have required the presence of some sort of a key device on Kirk's person. But the actual encryption is something that remains hidden from the audience...

Considering that future computers are extremely good at deciphering alien languages at breakneck speed, one might be tempted to think that all encryption systems based on a fictional language (in essence, all the encryption systems so far used by mankind) would be way too easy to break. Some sort of a conceptual change would have to be introduced: self-protecting codes, perhaps, active software entities that will adapt to breaking attempts and will secure themselves against the attack. Or codes that aren't mere abstractions, but make use of physical elements (similar to the quantum-entanglement code concepts being thrown around nowadays).

That said, Spock's father was credited with inventing or otherwise devising what was called a "satchel code" in the novel Spock's World. This supposedly was a practicable means of encryption at the time, not merely a meaningless hobby project for the Ambassador. Make of that whatever you will; for all we know it might refer to an active defensive software around the message, or to a physical package around the abstract message content.

Timo Saloniemi

Encryption cannot be broken no matter how fast the computer as it is an asymmetric process.

Right now I can think of a simple substitution encryption which cannot be broken even by a super computer for at least 10 years.

I am almost tempted to put a encrypted message below to prove it but i am so confident of myself i will not want to waste anyone time or cpu cycles.

 

[jefferiestubes8]

Encryption cannot be broken no matter how fast the computer as it is an asymmetric process.

noknowes you are speaking of public key-private key cryptography?

public key-private key cryptography is use of asymmetric key algorithms because the key used to encrypt a message is not the same as the key used to decrypt it. Each user has a pair of cryptographic keys — a public key and a private key. The private key is kept secret, while the public key may be widely distributed.

or

Proofs of security for asymmetric key cryptography therefore hold only with respect to computationally-limited adversaries, and can give guarantees (relative to particular mathematical assumptions) of the form "the scheme cannot be broken using a desktop computer in 1000 years", or "this algorithm is secure if no improved method of (for instance, integer factoring) is found".

http://en.wikipedia.org/wiki/Asymmetric_encryption

I'd like to ask if you think public key-private key cryptography will be in-use in 50,100,200 years from now?

 

[Crazy Eddie]

Even in the case of private-key, supercomputers have the advantage of the "brute force" approach, which--given a smart enough computer--is destined to eventually work.

Of course, this might be why people like Data prefer fractal encryption. The brute force method wouldn't work if the code is programmed to turn back false positives with each level of decryption. Arguably, all good algorithms in the 24th century use a similar method.

 

[Myasishchev]

Even in the case of private-key, supercomputers have the advantage of the "brute force" approach, which--given a smart enough computer--is destined to eventually work.

I'm pretty sure for numbers of sufficient size, the problem's intractable with classical computation. Physical constraints like energy requirements and time exceed what the whole universe has at its disposal. (Well, at least the time, I'm not looking up how much energy there is in the universe, but the point is that even the lower bound on the energy needed to factor large numbers is ridiculously large).

 

[Crazy Eddie]

Even in the case of private-key, supercomputers have the advantage of the "brute force" approach, which--given a smart enough computer--is destined to eventually work.

I'm pretty sure for numbers of sufficient size, the problem's intractable with classical computation. Physical constraints like energy requirements and time exceed what the whole universe has at its disposal. (Well, at least the time, I'm not looking up how much energy there is in the universe, but the point is that even the lower bound on the energy needed to factor large numbers is ridiculously large).

Yeah, but you'd have to assume some kind of arms race between cryptographic computers that necessarily requires encryption methods are only slightly ahead of decryption methods. The best algorithm in the world is useless if no computer in the universe is powerful enough to implement it.

 

[Romulan_spy]

In "Unification", Data was trying to break Romulan encryption. He says something about x layers of encryption and needing Spock's help. Does anyone remember what he says exactly? That episode gave the impression that Star Trek has some very sophisticated encryption keys, considering that an android like Data had difficulty breaking all of them.

 

[noknowes]

In "Unification", Data was trying to break Romulan encryption. He says something about x layers of encryption and needing Spock's help. Does anyone remember what he says exactly? That episode gave the impression that Star Trek has some very sophisticated encryption keys, considering that an android like Data had difficulty breaking all of them.

A very simple solution is double or triple encryption.

you encrypt something.then you encrypt the encrypted again!!!


There is no limit to how many times you can re-encrypt.


each time you re-encrypt the computer power required to break it rises as the power of the encrpytion.

thus encrypted twice it would rise by n+3 2+3 squared=25 times by 3 then =3+25 =29 cubed!!!

The process is assymetric.

 

[Cid Highwind]

A very simple solution is double or triple encryption.

you encrypt something.then you encrypt the encrypted again!!!

Yes! I'm currently using ROT13 encryption twice, on all my posts here!

I plan to change, doing it four times for double security!

 

[jefferiestubes8]

double encryption

A very simple solution is double or triple encryption.

The only argument that can be made for multiple layers of encryption is when you're using different algorithms.
Suppose you encrypt your message in AES, then in Serpent and finally in CAST. Now suppose that AES got broken or significantly weakened. They won't even know that there's an AES message in there.

SOURCE

 

[USS KG5]

A very simple solution is double or triple encryption.

you encrypt something.then you encrypt the encrypted again!!!

Yes! I'm currently using ROT13 encryption twice, on all my posts here!

I plan to change, doing it four times for double security!

It isn't working, I can read all of them

 

[Cid Highwind]

It isn't working, I can read all of them

Well, well... I'm sure by now it's illegal to break encryption, no matter how pathetic it is - so, have fun in jail!

Seriously, though: multiple encryption is not as secure as it might sound at first. For some encryption algorithms, encrypting with a key A, then reencrypting with a key B, may lead to the same ciphertext as if encrypting just once, using a third key C.

A trivial example of that is a cyclic rotation of all letters, which is what I referred to before. If I'm rotating by +3 ("A" becomes "D") and then rotating again by +2 ("D" becomes "F"), the resulting ciphertext will be the same as if I rotated by +5 directly.

Even for non-trivial encryption algorithms, where this might not apply, multiple encryption does not necessarily lead to an exponentially increased security. The so-called "Meet-in-the-middle-attack", for example, allows to break double encryption in just twice the time as single encryption (instead of the expected time^2).