Break the WWII pigeon code for bitcoin

[Micon]

http://www.cnn.com/2012/11/23/world/europe/uk-wwii-pigeon-mystery/index.html

Why can't anyone break this code?  they claim it might have involved a "one time pad"

uh... can't we super-compute the code on out of there real quick?

would be cool if bitcoiners did it... would get the community very positive attention.  Preferably a female cryptographer or a young kid.  That would be perfect for the media. We just *have* to have bad-ass cryptography experts combing these forums... (I am not one of them)

[1715297528]

1715297528

[1715297528]

1715297528

[1715297528]

1715297528

[jim618]

One time pads are similar to XORing a pre agreed random text with your message text to create cipher text.
If you do not have the pre agreed random text they are unbreakable.

Given some one time pad cipher text you can 'decode' it to any message you like of the same length by preparing the random text to XOR it with.

According to Wikipedia they were known about in 1882 so would have been in common use in WWII.

[greyhawk]

(I am not one of them)

That is very very obvious. Breaking a true OTP encrypted message is mathematically impossible no matter how much computer effort you throw at it.

[Remember remember the 5th of November]

Looks like a serial key to me.

[greyhawk]

Looks like a serial key to me.

Put it in Steam and see if it's viral marketing for a game. 

[Phinnaeus Gage]

AOAKN HVPKD FNFJW YIDDC
RQXSR DJHFP GOVFN MIAPX
PABUZ WYYNP CMPNW HJRZH
NLXKG MEMKK ONOIB AKEEQ
WAOTA RBQRH DJOFM TPZEH
LKXGH RGGHT JRZCQ FNKTQ
KLDTS FQIRW AOAKN 27 1525/6

I'm almost certain this was a time traveler carrier pigeon. MIAPX = Matthews Asia Growth Fund

For S&G, I searched DYKEG and foung the following: http://thenextweb.com/insider/2012/11/23/after-weeks-of-trying-uk-cryptographers-fail-to-crack-world-war-ii-code-found-on-dead-pigeon/

Kris  Massively • 3 days ago −

No, he is right, there are 27 groups. All he missed is what the /6 represented. Use the "DD" "YY" "KK" "EE" "GG" as line delimiters, string the groups together to make a line and you get 6 lines total, results in:

AOAKN

HVPKDFNFJWYI D
D CRQXSRDJHFPGOVFNMIAPXPABUZW Y
Y NPCMPNWHJRZHNLXKGMEM K
K ONOIBAK E
E QWAOTARBQRHDJOFMTPZEHLKXGHR G
G HTJRZCQFNKTQKLDTSFQIRW

AOAKN 27 1525/6

AOAKN could perhaps be combined with DYKEG somehow for the cipher...

I'm going to ask Maria to work on this with her ASICs.

[greyhawk]

K ONOIBAK E

That's interesting. Wasn't the pigeon found in a chimney?

Let's play with that string a bit.

O NO I BAKE

I think you're on to something.

[Micon]

If listening to MC Front-a-lot's Defcon 2012 "Secrets of the Future" more than twice I feel I am now qualified to comment on one-time pads. 

http://www.youtube.com/watch?v=ZP8VghUuX7M

Maybe it's just this stupid human brain in my head, but I sort of refuse to accept


AOAKN HVPKD FNFJW YIDDC
RQXSR DJHFP GOVFN MIAPX
PABUZ WYYNP CMPNW HJRZH
NLXKG MEMKK ONOIB AKEEQ
WAOTA RBQRH DJOFM TPZEH
LKXGH RGGHT JRZCQ FNKTQ
KLDTS FQIRW AOAKN 27 1525/6


that is totally unhackable.  It will likely need a savvy human or 2 with the aid of a savvy computer or 2 and try and bunch of top-level shit that the computers can churn on for a hot minute.  I.e. first test that makes sense to me is "what if those 5 letter groupings each represent one letter or number"  code the nuts out of a program to try many different odd ways to add up those letters / divide by  1 then 2 then 3 then take the last digit and match it against a decending alpha-numeric list and see if it spells "COORDINATES 12"W 67"N and try obviously a large number of schemes, a large number of iterations within each scheme... might be a cool incomplete information computing task...

and I hereby fully accept what appears to be the case here and I'm a stupid human who cannot wrap his head around 1xPad cryptography and Frontalot lied to me and said a children's speak-n-spell could crack it in 2025.

[greyhawk]

I.e. first test that makes sense to me is "what if those 5 letter groupings each represent one letter or number"  code the nuts out of a program to try many different odd ways to add up those letters / divide by  1 then 2 then 3 then take the last digit and match it against a decending alpha-numeric list and see if it spells "COORDINATES 12"W 67"N

That's exactly the beauty of One Time Pads. Without the Pad you can theoretically arrive at any plaintext you wish from the crypto text via cryptanalysis without ever knowing if the plaintext is correct or something you want it to be like

STUCK IN CHIMNEY PLEASE SEND HELP

[AndrewBUD]

Please send solvent

Glued hand to self

Please hurry


Ouch

[Phinnaeus Gage]

One thing is for sure, for being a short message, all the letters of the English alphabet are used.

~Bruno K~

[Phinnaeus Gage]

AOAKN HVPKD FNFJW YIDDC
RQXSR DJHFP GOVFN MIAPX
PABUZ WYYNP CMPNW HJRZH
NLXKG MEMKK ONOIB AKEEQ
WAOTA RBQRH DJOFM TPZEH
LKXGH RGGHT JRZCQ FNKTQ
KLDTS FQIRW AOAKN

Here's the letter distribution count:

A: 9
B: 3
C: 3
D: 6
E: 4
F: 7
G: 5
H: 8
I: 4
J: 5
K: 10
L: 3
M: 5
N: 9
O: 7
P: 7
Q: 6
R: 8
S: 2
T: 5
U: 1
V: 2
W: 5
X: 4
Y: 3
Z: 4

And the count by frequency:

K: 10

A: 9
N: 9

H: 8
R: 8

F: 7
O: 7
P: 7

D: 6
Q: 6

G: 5
J: 5
M: 5
T: 5
W: 5

E: 4
I: 4
X: 4
Z: 4

B: 3
C: 3
L: 3
Y: 3

S: 2
V: 2

U: 1

Now, it looks pretty simple to solve.

K = Contact
AN = Airborne Network (or Army/Navy)
HR = Human Resource/Relations; Hand Receipt(s); Humanitarian Relief
FOP = Flight Operations Plan
DQ = Data Quest
G = Grid           J = Jammed
MTW = Major Theater of War
EI = End Item               X = Exercise       Z = Zodiac
BCL = Battlefield Coordination Line           Y = Y-Axis (Yankee Time Zone)
SV = Sniper Variant
U = UNCLASSIFIED; Unrestricted

[Phinnaeus Gage]

My apologies, but this is an ego bump. I'm curious to read if my last post has merit. The final aspect does need a tad more polishing though, but I feel it's close.

Thanks, all.

~Bruno K~

[DarkHyudrA]

"T" could be Tank or Tiger, the so called heavy-armored tank, as P could be the Panzer?
"D" could be Division?

[yogi]

It looks enigma encrypted. But, that is to obvious and a brute force attack is likely to have already been tried.

Maybe it's a list of enigma codes?

[Phinnaeus Gage]

I think this pigeon deserves the Dickin Medal: http://en.wikipedia.org/wiki/Dickin_Medal

[DarkHyudrA]

I think this pigeon deserves the Dickin Medal: http://en.wikipedia.org/wiki/Dickin_Medal

He delivered his/her message to deserve the medal?

[caffeinewriter]

If we wanted to "super-compute" this, and it is a one time pad, you have a better chance of cracking a specific Bitcoin key-pair.

AOAKN HVPKD FNFJW YIDDC
RQXSR DJHFP GOVFN MIAPX
PABUZ WYYNP CMPNW HJRZH
NLXKG MEMKK ONOIB AKEEQ
WAOTA RBQRH DJOFM TPZEH
LKXGH RGGHT JRZCQ FNKTQ
KLDTS FQIRW AOAKN

Here's the letter distribution count:

A: 9
B: 3
C: 3
D: 6
E: 4
F: 7
G: 5
H: 8
I: 4
J: 5
K: 10
L: 3
M: 5
N: 9
O: 7
P: 7
Q: 6
R: 8
S: 2
T: 5
U: 1
V: 2
W: 5
X: 4
Y: 3
Z: 4

And the count by frequency:

K: 10

A: 9
N: 9

H: 8
R: 8

F: 7
O: 7
P: 7

D: 6
Q: 6

G: 5
J: 5
M: 5
T: 5
W: 5

E: 4
I: 4
X: 4
Z: 4

B: 3
C: 3
L: 3
Y: 3

S: 2
V: 2

U: 1

Now, it looks pretty simple to solve.

K = Contact
AN = Airborne Network (or Army/Navy)
HR = Human Resource/Relations; Hand Receipt(s); Humanitarian Relief
FOP = Flight Operations Plan
DQ = Data Quest
G = Grid           J = Jammed
MTW = Major Theater of War
EI = End Item               X = Exercise       Z = Zodiac
BCL = Battlefield Coordination Line           Y = Y-Axis (Yankee Time Zone)
SV = Sniper Variant
U = UNCLASSIFIED; Unrestricted

Quite an interesting analysis. I'd love to see the context this pigeon code was intercept in to verify your theory to see if it's viable.

[greyhawk]

It looks enigma encrypted.

No it doesn't. Enigma encrypted messages had a header like this:

[Phinnaeus Gage]

With my theory, where the letters are doubled may have no significance to the code. Also, has anybody considered if the blue ink use to pen it is relevant?

[P_Shep]

PG, you're not getting it. It's totally unbreakable. You can do all the analysis in the world, using all the computers in the world and come up with nothing.

http://en.wikipedia.org/wiki/One-time_pad

[mintymark]

If its a one time pad, its probably unbreakable. But the big problem with one time pads is that as a spy, if you are found with one, that pretty much proves it, so there are good reasons why a one time pad might NOT have been used.

What altertatives are there? Vigenère cipher seems an obvious possibility and would have been hard to crack if the key was long. The key would have been a quotation, carried in the head and could easily be long. Encryption can practically be done in the head by an experianced operative and fast, or with the aid of a simple nodepad if not. To crack it you should look for repeated sequences - these would be simple words encrypted on a given multiple of the key length which end up encrypted the same way. But there are not any and that may just mean the key is long.

Enigma? Dont forget that the British did not use Enigma, the Germans did, so this seems unlikely.

More historical context would help here. Exactly what was the norm for spies in 2WW Germany to use at this time? And could would please have a list of all the one time pads? How many are there, could we just brute force them? (If we had that?)

Could it be a book Cipher of sorts?

[caffeinewriter]

If its a one time pad, its probably unbreakable. But the big problem with one time pads is that as a spy, if you are found with one, that pretty much proves it, so there are good reasons why a one time pad might NOT have been used.

What altertatives are there? Vigenère cipher seems an obvious possibility and would have been hard to crack if the key was long. The key would have been a quotation, carried in the head and could easily be long. Encryption can practically be done in the head by an experianced operative and fast, or with the aid of a simple nodepad if not. To crack it you should look for repeated sequences - these would be simple words encrypted on a given multiple of the key length which end up encrypted the same way. But there are not any and that may just mean the key is long.

Enigma? Dont forget that the British did not use Enigma, the Germans did, so this seems unlikely.

More historical context would help here. Exactly what was the norm for spies in 2WW Germany to use at this time? And could would please have a list of all the one time pads? How many are there, could we just brute force them? (If we had that?)

Could it be a book Cipher of sorts?

In a way, an OTP is basically a book cipher. In other news:

Carriers of Secrets: Pigeon on Trial for Espionage

One time pads could be difficult, since they'd need to make sure the key was safely delivered as well, since there's not really an easy way to transfer large keys over long distance. Theoretically, another uncracked message could actually turn out to be a key, but it's highly unlikely that both were intercepted. If they're using shorthand as well, that makes the decryption process even more tricky, and renders cryptanalysis essentially useless.

In cryptography, the one-time pad (OTP) is a type of encryption which has been proven to be impossible to crack if used correctly.

Please note, it's not theoretically uncrackable, it's proven to be uncrackable. (If used correctly.) You'd have to find a chink in the German's process to make any decryption attempt viable.

To continue the example from above, suppose Eve intercepts Alice's ciphertext: "EQNVZ". If Eve had infinite computing power, she would instantly find that the key "XMCKL" would produce the plaintext "HELLO", but she would also find that the key "TQURI" would produce the plaintext "LATER", an equally plausible message:

As you can see, this is letter frequency, but if there's a shorthand code as well, letter frequency goes out the window. For example, trps a rhn, rt could mean troops at the rhine, retreat. However, word frequency get's significantly more difficult since it only uses some letters from the original word. It would be even more difficult should they have some completely off topic shorthand, such as gziek mat mforkan, flta could mean the exact same thing provided the receiver had knowledge on how to use it. Also, you'd have to figure out how the key was transmitted as well.

[farlack]

The worlds best could not do it I'm sure they went over all these theories. Not sure you guys are understanding the one time thing.


A coded book, with the answers to the coded message. Used 1 time, then destroyed.

Unless you have one of those 2 books, you're not going anywhere.


ASDW#$FGDXCV in the coded book can mean, Allied victory at this place, soviet retreat here.

You'll never crack that.

[caffeinewriter]

Well, regardless, we should stop this carrion. (Insert collective groans here)

[TheButterZone]

Squab!

[Phinnaeus Gage]

The worlds best could not do it I'm sure they went over all these theories. Not sure you guys are understanding the one time thing.

A coded book, with the answers to the coded message. Used 1 time, then destroyed.

Unless you have one of those 2 books, you're not going anywhere.

ASDW#$FGDXCV in the coded book can mean, Allied victory at this place, soviet retreat here.

You'll never crack that.

How do we know for certain a OTP was used? If I read correctly, there were millions of messages sent via pigeons during WWII. To use a OTP, don't you have to sent a pigeon first with the key prior to sending the crypto message? If that's the case, then what happens if the other end doesn't get the key but only the message?

FYI: Did you know that some prisoners are using pigeons to get drugs onto the grounds? Who here is working on Silk Air?

~Bruno K~

[greyhawk]

No, the recipient already has his copy of the OTP way way before any pigeons are ever sent.

[caffeinewriter]

No, the recipient already has his copy of the OTP way way before any pigeons are ever sent.

Unless they had a whole library of them, it isn't technically a one-time pad.

[Phinnaeus Gage]

No, the recipient already has his copy of the OTP way way before any pigeons are ever sent.

Unless they had a whole library of them, it isn't technically a one-time pad.

If that's the case, then the numbers at the end of the message may be which pad to use to decode it.

Would somebody be kind as to explain to me (or others interested) as to why the length of the words(?) are all five letters long?

~Bruno K~

[SgtSpike]

No, the recipient already has his copy of the OTP way way before any pigeons are ever sent.

Unless they had a whole library of them, it isn't technically a one-time pad.

If that's the case, then the numbers at the end of the message may be which pad to use to decode it.

Would somebody be kind as to explain to me (or others interested) as to why the length of the words(?) are all five letters long?

~Bruno K~

My guess would be that spaces are removed in such a cipher to make it that much more difficult for an enemy to crack.  Once decoded, it is easy enough to figure out where the spaces should go.

OR, one of the letters represents a space.

They're grouped into 5's to be easier to read/decode without losing one's place.

[John (John K.)]

I just installed windows XP using the first row.


(I admit I plagiarized this =P)

[farlack]

The worlds best could not do it I'm sure they went over all these theories. Not sure you guys are understanding the one time thing.

A coded book, with the answers to the coded message. Used 1 time, then destroyed.

Unless you have one of those 2 books, you're not going anywhere.

ASDW#$FGDXCV in the coded book can mean, Allied victory at this place, soviet retreat here.

You'll never crack that.

How do we know for certain a OTP was used? If I read correctly, there were millions of messages sent via pigeons during WWII. To use a OTP, don't you have to sent a pigeon first with the key prior to sending the crypto message? If that's the case, then what happens if the other end doesn't get the key but only the message?

FYI: Did you know that some prisoners are using pigeons to get drugs onto the grounds? Who here is working on Silk Air?

~Bruno K~

AOAKN HVPKD FNFJW YIDDC
RQXSR DJHFP GOVFN MIAPX
PABUZ WYYNP CMPNW HJRZH
NLXKG MEMKK ONOIB AKEEQ
WAOTA RBQRH DJOFM TPZEH
LKXGH RGGHT JRZCQ FNKTQ
KLDTS FQIRW AOAKN 27 1525/6

Possibly, but the numbers 1525/6 could have something to do with it? Like multiply 1525*6 1525+6 1525/6 etc  then change the text around depending on the results.

But I would have to assume that its more like 1525/6 is a section in a decoder book, that one person has, and once the letter is received, they destroy both. I wouldn't know I only learned about one time pads when I read the news article when it came out a few weeks ago hah..
Apparently these notes, and books, are highly flammable so if they were over-run, just give it some heat and its up in flames.

[TheButterZone]

Nitrocellulose?

[J-Norm]

Some of these were broken despite the fact the one time pad systems cannot be "broken" persay.

The combination of re-using keys, stolen code books and using the same prefix at the start of certain daily messages(like weather) they did manage to crack several messages using primitive purpose build computers.

If you can get access to all of the stolen/derived code data then you could run it all through and see if any decodes it. Otherwise forget about cracking it, you will get quotes from the bible before getting the real data.

All of this assumed a one-time pad. This does not look like a one time pad due to the uneven character distribution. Then again with such a small sampling such unevenness can happen in truly random data.

The uneven distribution could also be related to the fact the the "random" one time pads were often generated by mashing a keyboard. People tend to bias towards the middle of a keyboard, if the message had repeating characters then this bias would show through.

[greyhawk]

Some of these were broken despite the fact the one time pad systems cannot be "broken" persay.

The combination of re-using keys, stolen code books and using the same prefix at the start of certain daily messages(like weather) they did manage to crack several messages using primitive purpose build computers.

Yeah, but to do that you need a lot more cyphertext and a reasonable estimation of what parts of the plaintext would be.

[J-Norm]

No, the recipient already has his copy of the OTP way way before any pigeons are ever sent.

Unless they had a whole library of them, it isn't technically a one-time pad.

The common form was a small book with the pages glued together at the sizes made of nitro paper. It burns completely and quickly. You take pages off the take when you used up your keys.

[J-Norm]

Yeah, but to do that you need a lot more cyphertext and a reasonable estimation of what parts of the plaintext would be.

Yes, modern computers would not really help much. The mechanical computers they made could keep up with the cryptographers guess lists. A team of very smart people pouring over information about the time/place with acccess to reams of prior code data will get just a few of these codes regardless of computer power.

I doubt this will ever be broken because even if the key can be found/derived via cribs the work involved would be too much. With a provably unbreakable code system you can only do side channel attacks or attack flaws in the implementation.

[Phinnaeus Gage]

Did anybody notice that this message was sent twice? (see lower right of image)

Check out this link: http://www.ciphermysteries.com/2012/11/24/dead-ww2-cipher-pigeon-timeline

[farlack]

    AOAKN - Artillery Observer At "K" Sector, Normandy
    HVPKD - Have Panzers Know Directions
    FNFJW - Final Note [confirming] Found Jerry's Whereabouts
    DJHFP - Determined Jerry's Headquarters Front Posts
    CMPNW - Counter Measures [against] Panzers Not Working
    PABLIZ - Panzer Attack - Blitz
    KLDTS - Know [where] Local Dispatch Station
    27 / 1526 / 6 - June 27th, 1526 hours


http://www.bbc.co.uk/news/uk-20749632

[Phinnaeus Gage]

    AOAKN - Artillery Observer At "K" Sector, Normandy
    HVPKD - Have Panzers Know Directions
    FNFJW - Final Note [confirming] Found Jerry's Whereabouts
    DJHFP - Determined Jerry's Headquarters Front Posts
    CMPNW - Counter Measures [against] Panzers Not Working
    PABLIZ - Panzer Attack - Blitz
    KLDTS - Know [where] Local Dispatch Station
    27 / 1526 / 6 - June 27th, 1526 hours


http://www.bbc.co.uk/news/uk-20749632

Beside PABLIZ having six letters, while all the others have five (should be PABUZ), coupled with there not being any rhyme or reason for picking only those seven groupings, I would say this ain't it.

~Bruno K~

[caffeinewriter]

Cracked by a Canadian:

http://www.cbc.ca/news/technology/story/2012/12/18/toornto-carrier-pigeon.html

It supposedly says "artillery observer/spotter at K-sector Normandy"

[cedivad]

This dosent make any sense.
I could have invented a solution, too. After all in English there are more acronimous than words.

[Phinnaeus Gage]

Cracked by a Canadian:

http://www.cbc.ca/news/technology/story/2012/12/18/toornto-carrier-pigeon.html

It supposedly says "artillery observer/spotter at K-sector Normandy"

Highly trained, incredibly skilled carrier pigeons (and at least one not operating on all thrusters, thus opting to refuel, causing its demise) delivered messages between allied troops in Europe.

My crackin' skills produced a better translation, and I have a hole in my head.

AOAKN HVPKD FNFJW YIDDC
RQXSR DJHFP GOVFN MIAPX
PABUZ WYYNP CMPNW HJRZH
NLXKG MEMKK ONOIB AKEEQ
WAOTA RBQRH DJOFM TPZEH
LKXGH RGGHT JRZCQ FNKTQ
KLDTS FQIRW AOAKN

Here's the letter distribution count:

A: 9
B: 3
C: 3
D: 6
E: 4
F: 7
G: 5
H: 8
I: 4
J: 5
K: 10
L: 3
M: 5
N: 9
O: 7
P: 7
Q: 6
R: 8
S: 2
T: 5
U: 1
V: 2
W: 5
X: 4
Y: 3
Z: 4

And the count by frequency:

K: 10

A: 9
N: 9

H: 8
R: 8

F: 7
O: 7
P: 7

D: 6
Q: 6

G: 5
J: 5
M: 5
T: 5
W: 5

E: 4
I: 4
X: 4
Z: 4

B: 3
C: 3
L: 3
Y: 3

S: 2
V: 2

U: 1

Now, it looks pretty simple to solve.

K = Contact
AN = Airborne Network (or Army/Navy)
HR = Human Resource/Relations; Hand Receipt(s); Humanitarian Relief
FOP = Flight Operations Plan
DQ = Data Quest
G = Grid           J = Jammed
MTW = Major Theater of War
EI = End Item               X = Exercise       Z = Zodiac
BCL = Battlefield Coordination Line           Y = Y-Axis (Yankee Time Zone)
SV = Sniper Variant
U = UNCLASSIFIED; Unrestricted

[Phinnaeus Gage]

Expert dismisses claims WWII pigeon code has been cracked

I don't believe anybody has offered a reward to the first person who cracks it. What if some Bitcoin entity did such that via a press release? Wouldn't that bring awareness to Bitcoin, but in a better light than the Mitt Romney episode? I'm thinking Bitcoin 100.

Better yet, what if a whole slew of Bitcoin entities got on board, each offering a BTC100 reward/prize to the first person who cracks the Pigeon Code. Bitcoin Magazine could dedicate an entire article to such an endeavor, for it is crypto related. And every subsequent issue will have a paid ad offering up the reward, paid for equally by all the Bitcoin entities offering up the prize.

The list may look like the following:

Bitcoin 100
Butterfly Labs
Memory Dealers
Bitcoin Magazine
Bit Pay
Mt Gox
Seals with Clubs
Bit Brew
Bees Brothers
BitInstant
Casascius

Assuming 20 names on the list, that would equate to over a $25K prize, and the ad would only cost 1/20 the prevailing rate for each business on the list.

As long as there's at least 20 names on the list, there probably wouldn't be any problem with some advertisers bowing out every now and then. Then again, the list may grow, demanding a larger ad space, and less that each entity has to dole out to be placed on the list. And the list will be on the online version where each name is linked to their respected website. Looks to me like a win-win can't-lose proposition.

~Bruno K~