PDA

View Full Version : Enigma Codebook Tool



drdefcom
11-10-2005, 04:41 PM
For those who use the Enigma Simulator v4.2 (http://users.telenet.be/d.rijmenants/en/enigmasim.htm), a brandnew freeware Enigma Codebook Tool (http://users.telenet.be/d.rijmenants/en/codebook.htm) is now also available.

This tool can create codebooks for the 3-rotor Wehrmach/Luftwaffe Enigma, the 3-rotor Kriegsmarine M3, also called Funkschlussel M, and the 4-rotor Kriegsmarine M4. The program generates random settings, depending the model, for the eight different normal and two special Beta and Gamma rotors, for both wide and thin reflectors, ringsettings and plugboard connections.

It also includes a helpfile which explains the procedures to set up the Enigma and send messages http://forums.ubi.com/groupee_common/emoticons/icon_cool.gif

drdefcom
11-10-2005, 04:41 PM
For those who use the Enigma Simulator v4.2 (http://users.telenet.be/d.rijmenants/en/enigmasim.htm), a brandnew freeware Enigma Codebook Tool (http://users.telenet.be/d.rijmenants/en/codebook.htm) is now also available.

This tool can create codebooks for the 3-rotor Wehrmach/Luftwaffe Enigma, the 3-rotor Kriegsmarine M3, also called Funkschlussel M, and the 4-rotor Kriegsmarine M4. The program generates random settings, depending the model, for the eight different normal and two special Beta and Gamma rotors, for both wide and thin reflectors, ringsettings and plugboard connections.

It also includes a helpfile which explains the procedures to set up the Enigma and send messages http://forums.ubi.com/groupee_common/emoticons/icon_cool.gif

p-11.cAce
11-10-2005, 06:01 PM
Thanks for the find! I love my ENIGMA http://forums.ubi.com/groupee_common/emoticons/icon_biggrin.gif

p-11.cAce
11-10-2005, 07:20 PM
I found this information to be very helpful in setting up my ENIGMA and understanding the "preamble" of the messages:
http://www.pbs.org/wgbh/nova/decoding/enigma2.html

drdefcom
11-11-2005, 10:45 AM
Hi p-11,

I saw the article on Nova, but it only describes the unsafe pré-war methode with double encoded trigram. In the Codebook Tool helpfile, the wartime methode of random start and random message key is explained. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Celeon999
11-11-2005, 01:37 PM
http://forums.ubi.com/images/smilies/11.gif Thanks for the info

Celeon999
11-11-2005, 01:58 PM
Some friends of mine use the Enigma to communicate via radio. http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

They will be glad to get this program although they mainly use the Sonderschlüssel 100 (Special Key 100) mainly used by the "Abwehr" (german secret service) that Bletchley Park never managed to decode. http://forums.ubi.com/groupee_common/emoticons/icon_smile.gif

W.Irving
11-11-2005, 03:16 PM
Luckily Abwehr were the good guys. http://forums.ubi.com/images/smilies/16x16_smiley-indifferent.gif

Jose.MaC
11-12-2005, 11:00 AM
by the way, what are the chances of breaking those codes? I mean, with nowadays computers, how long would take to someone using a enigma simulator to break a message? Since our PC are able to work at speeds over one megahertz, maybe single coded messages are breakable just in a day, and triple coded messages are breakable in less than a week, even using poor decode algorithms.

Celeon999
11-12-2005, 01:52 PM
The 4 rotor Enigma has exactly 17.298.883.602.000 possible configurations.

Today it would be easy to create a program that could go through all combinations in short time.

But....

First you need to know how the internal wiring and mechanism looks like. You can find information about this in the internet but its still a very complex thing to understand.

Then you need to create a program that is based on this alghorithms. So you have to be a pc programer and a diplom mathematican in one person.

(Propably like the persons that created the simulator http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif )

And then there is still one task to do that no computer can do :

The computer shows you all 17.298.883.602.000 decoded variants but it cant read the messages and so it cant know which one is an decoded message and which are the 17.298.883.601.999 other messages that contain only unreadable ****.

This can only be done by human beeing. And to look through all 17.298.883.602.000 pages ......

That could take some time .... http://forums.ubi.com/images/smilies/16x16_smiley-very-happy.gif

http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif

Now you could combine the encoder with a program for grammatical correction like the one MS Word uses but ... there is still another problem...

You dont know in which language the message is written http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif


You see. Its makable .

But it requiers an unbelievable amount of work and time. http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif

So you would need to draw the attention of someone on your messages that has access to the know how , a lot of free time, and alot of money. Like Bletchley Park in ww2. http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

So for a private person´s use its 100 % safe and you could even encode your bank account pin numbers or whatever you want print them out on 100 sheets of paper and throw them out of the window if you like. http://forums.ubi.com/groupee_common/emoticons/icon_biggrin.gif

Jose.MaC
11-12-2005, 07:18 PM
All right. 17.298.883.601.999 combinations.

Acording with this page, http://www.myke.com/enigma1.htm you may be able to decode in two minutes a enigma message with a relatively simply tool in a low end computer. Assuming that you've another utility to check every combination to erase clear noise (i.e.: combinations of nine consonants or six vowels) that doubles the time, you get cribed versions in 4 minutes. If you do an utility to check similarity of words with some idioms (german, english, spanish, french... you only need about 100 words of every idiom), you can do another crib that may use 20 minutes. So in about an hour, you can decode any enigma message... even if you're using a very poor decode algorithm. Of course, keeping messages short and using codes will make harder the decoding process.

Even if what this guy says is false, our computers can do more than 1.000.000.000 cicles per second, with 2 operations per cicle. So is about 2.000.000.000 cicles. Asuming that you need 1.000 cicles to do a whole combination, you need 30 computers working at 3 GHz to do all combinations in a day. Since 95% of messages must be done with only 60% of combinations (Gauss come handy here http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif) you only need 20 computers to do so.

Anyway, is out of question than usually Huff-Duff would prove more eficient than trying to break enigma messages.

drdefcom
11-13-2005, 03:31 AM
Hey guys, the updated version of the Codebook Tool (http://users.telenet.be/d.rijmenants/en/codebook.htm) is available, with nice sheet printout! http://forums.ubi.com/groupee_common/emoticons/icon_cool.gif http://forums.ubi.com/groupee_common/emoticons/icon_cool.gif btw, Jose.mac, it's not that simple, counting the clockcicles of your PC. Since there are always less then 250 chars in a message there are quiet some fifficulties to overcom. Expert in this matter is Frode Weierud (http://frode.home.cern.ch/frode/crypto/index.html) from CERN, he broke a large quantity if original messages with the CSG. Must take a peek at his site http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

drdefcom
11-13-2005, 03:44 AM
Oh Celeon999, you're a bit wrong about the combination. The mathematical correct (theoretical) possible combinations of an 4-rotor machine are 23,276,989,683,567,292,244,023,724,793,447,
227,628,130,289,261,173,376,992,586,381,072,041,
865,764,882,821,864,156,921,211,571,619,366,980,73 4,115,
647,633,344,328,661,729,280,000,000,000,000,000 http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif

which is approximately 2 Ӕ 10E145 (source NSA (http://www.nsa.gov/publications/publi00004.cfm) )

However, the practical number of combination is 10E23 (which is still 100,000,000,000,000,000,000,000 http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif) This means one must apply cryptographic analysis, while a brute force attact, as you suggest, would take far to long. See, it's not that simple! http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Celeon999
11-13-2005, 04:28 AM
23,276,989,683,567,292,244,023,724,793,447,
227,628,130,289,261,173,376,992,586,381,072,041,
865,764,882,821,864,156,921,211,571,619,366,980,73 4,115,
647,633,344,328,661,729,280,000,000,000,000,000

http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif

I had my number from wikipedia.

Does this number above already contain the possibility of an free programable D Reflector that may be used ? http://forums.ubi.com/images/smilies/blink.gif


Its cool that the screen of the Tool is made red like the real codebooks were. At least the codebooks used on ships and u-boats were made so.

They were printed on special "Rotdruck" paper that was soluble in water.

So if an u-boat or ship was sunk the codebooks were automatically destroyed as soon as they came in contact with water. http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif

drdefcom
11-13-2005, 05:11 AM
Actually, the codebooks on wikipedia are the Kenngruppenhefte, no enigma keys. Normally, the codes for 4-rotor machines were stenciled on two normal papiers, one with rotor settings (every TWO days) and one sheet for plugboard settings (every day). For more info take a peek at Frodes (http://frode.home.cern.ch/frode/crypto/index.html) pages (TRITON keys (http://frode.home.cern.ch/frode/crypto/Triton.pdf) ) Maybe I will update my program to make the M4 sheets look like that, but this would require two seperate sheets. Still thinking about that change. Maybe, the way i wrote it now is more practical, and shows a fairly realistic combination of 3-and4 rotor codes (did you try out the v2.0 of the tool, with better printout? first uninstall the old version!) If you have any suggestions or ideas, you can always mail me through my website (http://users.telenet.be/d.rijmenants/index.htm) http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

Celeon999
11-13-2005, 06:02 AM
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">(did you try out the v2.0 of the tool, with better printout? </div></BLOCKQUOTE>

Yep. Just installed it http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif http://forums.ubi.com/images/smilies/25.gif

Jose.MaC
11-13-2005, 09:25 AM
Now, I wonder how they were able to break a single code... they simply lacked of the brute force to do so!

Celeon999
11-13-2005, 09:50 AM
<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">Now, I wonder how they were able to break a single code... they simply lacked of the brute force to do so! </div></BLOCKQUOTE>

Well, Bletchley Park organized themselves some Enigmas from different sources.

The Enigma was once used for commercial purposes so it wasnt hard to find some in the u.k. .

They also had help from polish cryptoanalysts that had a lot of experience in decoding Enigma messages.

That combined with captured Kriegsmarine Enigmas and Codebooks gave them the tools for building the decoding computers used for project ULTRA.

drdefcom
11-13-2005, 12:49 PM
They often used the technique of cribs. The Germance were in fact to diciplined http://forums.ubi.com/images/smilies/35.gif . They used often the same abbreviations, standard message formats, and things like "es lebe den fuhrer", combined with wath we call trafic analysis: knowing what could be in a message by finding out from who it is, and where its gooing to, in wich periode and tactical conditions. Bletchley park would guess pieces of a message. These were the famous cribs. Next, they would try to find those settings to make that little piece readable (done with a "bombe") and then decipher the rest. It's a bit like mini-brute-force, but first apply cryptanalysis to greatly reduce the number of possibilities. In the Atlantic battle, sometimes, due to lack of cribs, they used a technique called "gardening": they planted sea-mines on specific locations. U-boats would send a standard contact-message if they spotted those mines. So, if they captured a message, and it was triangulated in that area, they had new cribs to break the settings for that day. Breaking 4-rotor Enigmas could take up to 40 http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif days with a 3-rotor bombe, but the U-boats used the M4 often in compatible mode to communicate with the M3's from weatherships etc http://forums.ubi.com/images/smilies/51.gif . So, they could break them within 24 hrs with the bombes. Later, they build 4-rotor-bombes to crack the M4 faster. Don't forget, only in Bletchley, over 7000 http://forums.ubi.com/groupee_common/emoticons/icon_eek.gif people were occupied breaking messages... btw, they never attacked u-boats for the Enigma's (they already had enigma's in '39) , but only for the codebooks, to break into the Triton net. They never knew about the introduction of the M4, but discovered it by cryptanalysis, that a 4th rotor was used. The Britisch were actually 10 months totally blind http://forums.ubi.com/groupee_common/emoticons/icon_frown.gif Captured codebooks and hacking the kurzsignalheft codes (those msg's were also sent over enigma and gave them cribs) were the breakthrough. Forget the silly U571-like bull**** movies, that's hollywood-**** http://forums.ubi.com/images/smilies/51.gif

Celeon999
11-13-2005, 01:45 PM
@ drdefcom

Ive noticed that you included some historical messages for decipher training in the help section of your simulator. (Like the last transmission of the bismarck)

Is there a place where i can find more historical messages ?

It makes fun decoding them http://forums.ubi.com/images/smilies/16x16_smiley-wink.gif


Ah yes and a second question :

Ive noticed that the codebook sheets dont have an instruction for the starting position of the wheels ??

Im also interested in that programmable D reflector . "...rarely used..."

Do you have more information on this reflector ?

Are you going to implement it in another version of the simulator ?

drdefcom
11-14-2005, 01:55 AM
Original messages are very rare. Frode Weierud from Crypto Simulation Groups (CSG) has (i believe finished) broken about 500 messages. He got then of the widow from a Dutchmen, who got them from a German who didn't destroy them and wanted to preserve them. All kinds of stuff, from the Russian campagne to the Flossenburg concentration kamp messages. Some are published on Frodes site see Flossenburge (http://frode.home.cern.ch/frode/crypto/Flossenbuerg/flo.html) where you can find the originals and their settings. Many messages and theire cryptanalysis will be published in Cryptologia. Don't know what or when...

Your second question about the startposition: This was never determined in advance. Befor the war, they sent the encrypted messagekey twice, and the startposition came from the sheets. This was a big mistake http://forums.ubi.com/images/smilies/351.gif During the war, they used a random startposition and random message key (see codebook tool helpfile) The reason is obvious: before the war, all message keys were encrypted with the same startposition. This gave cryptanalists lots of data to procces (more enciphered stuff with the same key is more change to find a solution) During the war, they used each message key a new random startposition, so that the encipherment of all those message keys had nothing in common.

This is the main reason why the Polish codebreaker got blind http://forums.ubi.com/images/smilies/cry.gif in 1939, when the Germans changed to this new procedure. They relied on the pré-war double encrypted message key.

So, random startposition procedures limited the extensive use of a particular enigma setting, just as avoiding long messages (more than 250 chars was forbidden, and they devided the message in parts) http://forums.ubi.com/groupee_common/emoticons/icon_wink.gif

About the simulator, well, this one has took loooots of free time and no update are planned (unless there's a bug or so) The programmable 'D' was that rare that i don't have any paperwork on procedures or keysheets that use it. If you want to see lots of enigma stuff, you should peek at Tom Perera's museum (http://w1tp.com/enigma/)

Celeon999
11-14-2005, 02:54 AM
Thanks for the info http://forums.ubi.com/images/smilies/11.gif