Decoding Encryption with Permission...

[garys]

The German Enigma machine was captured from a Nazi U-boat. It was captured by chance, there was no special mission to capture it as depicted in the movie U-571. Ironically the German navy Enigma was not cracked untill after the war.

My recollection is a bit different. Nova had a documentary on this a few years back, so my memory might also be a bit fuzzy. The Enigma machines had either three or four cylinders. The early ones were three but when the Germans realized that the British had broken the codes for the U Boats, they developed the four cylinder machine. Which also was eventually broken using the computer as you describe.

Anyway, there is a fair amount of information out there. There's even a guy that travels to a lot of ham flea markets in the Northeast and sells DVDs with a documentary. He has some machines on his display table.

Gary

 

[SenderGreen]

My recollection is a bit different. Nova had a documentary on this a few years back, so my memory might also be a bit fuzzy. The Enigma machines had either three or four cylinders. The early ones were three but when the Germans realized that the British had broken the codes for the U Boats, they developed the four cylinder machine. Which also was eventually broken using the computer as you describe.

Anyway, there is a fair amount of information out there. There's even a guy that travels to a lot of ham flea markets in the Northeast and sells DVDs with a documentary. He has some machines on his display table.

Gary

That's the one I saw! I thought there was one that was never broken during the war, I thought it was the Marinekreig (Navy) Enigma. I do remember something about a code called "Fish" though. Maybe it was 3 cylinder on each machine Tx and Rx. I know the LON-DON, MAD-RID, HIT-LER, BER-LIN, thing is not my imagination (I'm not that clever).
I may need to check my facts, thanks for the feedback.

EDIT: My mistake "On June 23rd, 1941, Bletchley Park decoded a U-boat message that would save a convoy." Also it looks as if the code that I was thinking was broken due to operator laziness/repetition was FISH on the Lorenz machine.

EDIT: found this http://www.pbs.org/wgbh/nova/decoding/enigma.html
http://www.pbs.org/wgbh/nova/decoding/mind.html
Can't watch the episode online, but there is a transcript: http://www.pbs.org/wgbh/nova/transcripts/2615decoding.html

 

[poltergeisty]

You should read this guys blog.

http://rijmenants.blogspot.com/

more notably this. http://users.telenet.be/d.rijmenants/

It seems that to understand the encrypted past, one still still needs to decode it for the present understanding.

 

[netsmith]

How'd you do that ?

Perhaps I am looking at this a bit late, but Rick rfmobile, posted a file converting the .mp3 to a bitsream?

Can I ask how that was done? What software/tools did you use?

Would the next step be to decrypt that bitstream?

I am baffled ;-)

Cheers,

Casey

PS - I think it will take longer than a week to decode, I just wanted to try a stab at it

 

[SCPD]

Perhaps I am looking at this a bit late, but Rick rfmobile, posted a file converting the .mp3 to a bitsream?

Here 'tis in three steps - the first two are a lazy way to get around the MP3 coding ...

1. patch cable from headphone output to microphone input.
2. Windows media player to play the MP3 file (at moderate volume so as not to smoke the microphone input)

That allowed me to decode directly from the microphone input as if I was monitoring discriminator audio from a radio.

3. I threw together some code to slice the waveform into a 12000 bit per second stream. Bit phase is inferred from +'ve and -'ve transitions through zero with some damping for noise. Bit level was determined by mid-sampling (sampling the waveform at the time believed to be the middle of a bit).

I already had code to do this ... (for Motorola and EDACS trunking) ... so it only took minutes to throw this together.

-rick

 

[rescue161]

Rick, did you see that I put the whole key out there? Does that help at all? Maybe this will make it somewhat easier in the future for folks to decode my messages.

 

[SCPD]

Yes I did - and I have no idea what to do with it.

 

[netsmith]

Decryption - Take 1

Here is my attempt at decrypting the transmission. This is not the final answer yet, but one step in the process.

I ran the file that rfmobile created through a decryption program using the key
F729B03EA4158C6D

The file still needs to be run through a codec that can return the result of the audio...
Not sure how to do that...

All I did was run the entire file through the DES algorithm, if there is another point in the file that I should begin decryption let me know.


As I hope this thread has pointed out, "making the theoretical practical" is not so easy...

Cheers,

Casey

A DES Illustrated Spreadsheet,

http://www.crypticide.com/dropsafe/article/1497

SecureNet_2.Dat.zip (remove the .zip extension for a raw decryption of the SecureNet.DAT)

 

[rescue161]

Keep up the good work guys. I am unable to open the file that you sent. It keeps trying to open it in Media Player and then says it's invalid? What program should I use? I did take the ".zip" off.

 

[netsmith]

File Extension

Once you remove the .zip extension, you still don't have an audio file that can be played in windows media player,

what you have is a file that needs to be run through some sort of codec that can play the audio... I am not sure what codec Motorolla uses to compress and digitize the voice...

I wasn't sure how to do that, but thought some folks on this thread might.

If you open the file with a Hex editor you can see the raw binary.

As I mentioned in the previous thread, the file I posted, still needs some work, even then I am not sure if I started decrypting in the right position...

I'll keep working on getting an audio version of the file...

Cheers,

Casey

 

[N_Jay]

. . .I am not sure what codec Motorolla uses to compress and digitize the voice...
. . . .

12 KBPS CVSD

 

[netsmith]

File Conversion

I can't seem to get the file converted to that format. I was using a program called Vox Studio that converts audio file types.

However it only supports CVSD in 24kbps...

Any ideas on another program that would play that file??

Cheers,

Casey

 

[N_Jay]

I can't seem to get the file converted to that format. I was using a program called Vox Studio that converts audio file types.

However it only supports CVSD in 24kbps...

Any ideas on another program that would play that file??

Cheers,

Casey

CVSD is a simple algorithm.

Just write a codec for it. :lol:

 

[netsmith]

Decrytion - Take 1 ... A Bust ;-)

Either I missed the decryption or missed the codec conversion.

When I play that file... All I get is more static.

Oh well. I though it was at least worth a stab at it. We should be able to decrypt it given the key!!!

Cheers,

Casey

 

[rescue161]

Either I missed the decryption or missed the codec conversion.

When I play that file... All I get is more static.

Oh well. I though it was at least worth a stab at it. We should be able to decrypt it given the key!!!

Cheers,

Casey

Here is the key:

F729-B03E-A415-8C6D

Logical ID: 0161

 

[netsmith]

F729b03ea4158c6d

That was the key I used, I was just saying,

If DES is so breakable then someone should be able to crack your audio stream, since yougave us the key...

Obviously its harder than most people believe.

I'll keep trying ;-)

Cheers,

Casey

 

[netsmith]

Trasmission

The file that was create by rfmobile, was for 12kbps, is that the transmission rate, or the sampling rate for the codec?

Do we need to create another file with a different transmit speed?

Just curious.

Cheers,

Casey

 

[SCPD]

The decrypted result would be a stream of CVSD bits. You would have to write a program to transcode this to a WAVE audio file. Here are some reasons why a simple pass through a DES de-crypter won't work.

frame sync - DES operates on 64 bit blocks. How do you know that the first bit of that file is the first bit of a block?

bit endian - are these blocks transmitted as big or little endian?

CVSD format - is this 1 bit slope or 2 bits per slope change, or something else?

interleave - were the encrypted bits interleaved prior to transmission?

error correction - are some of the 12000 bits per second there to provide error correction?

additional data - notice Scott posted the logical ID with the DES key. Is additional ANI data mixed in with the voice data?

-rick

 

[netsmith]

I am definitely going to leave decryption to the experts ;-)

I have learned a ton as result of attempting.

On closer inspection it seems there were 3 distinct transmissions but looking at the gaps in the file.

But based on your advice, it is going to take a lot more time and someone way smarter than me to try to piece that together, even if I have the key!

It was a fun challenge and thanks rick for your posts and quick responses.

Like I said, I learned a ton about what it takes to try to decrypt DES, and learned alot about what I don't know... ;-)

Cheers,

Casey

 

[poltergeisty]

SecureNet_2.Dat.zip (remove the .zip extension for a raw decryption of the SecureNet.DAT)

Why not put a .dat file in a .zip file? :lol: http://en.wikipedia.org/wiki/CVSD