wannabescannist
Member
I'd be interested...though I probably wouldn't be able to crack it.
Decoding Encryption with Permission...
- Thread starter rescue161
- Start date
- Status
MattSR
Member
/thread revival.
I think we should try this competition again, only this time with P25 and something like, oh ADP encryption.
Anyone keen?
I think we should try this competition again, only this time with P25 and something like, oh ADP encryption.
Anyone keen?
I can do it again with P25 using DES, or AES, but don't have ADP on the radio that I'm using.
If we are going to give this another go we should first post the basic info for the various encryption systems. Namly the key size, if the output is ECB,CBC,OFB etc. DES-OFB is very popular but ADP is becomming quite common too.
- Joined
- May 7, 2004
- Messages
- 4,012
- Reaction score
- 132
Duh! What was I thinking! Doing that would change the bits, huh? :lol: :lol:
MattSR
Member
OK - I think P25 DES-OFB is ideal. Since I dont have $100k for a FPGA cluster lying around, and to make it possible on a normal PC, part of the key will need to be known.
I can give out the entire key and both encrypted and decrypted messages if need be.
I do have a lot of other projects going on right now as well, so it may be latter than sooner. It's easy for anyone to do however.
I do have a lot of other projects going on right now as well, so it may be latter than sooner. It's easy for anyone to do however.
Wirelessly posted (Mozilla/5.0 (iPhone; U; CPU iPhone OS 3_1_3 like Mac OS X; en-us) AppleWebKit/528.18 (KHTML, like Gecko) Version/4.0 Mobile/7E18 Safari/528.16)
Distributed computing... Anyone?
Distributed computing... Anyone?
That would help I would like to look at this using P25 des-ofb. There is already a decent linux based program that will decode regular unencrypted audio and play it on a sound card. All that would be needed would be to add in a routine to take the decoded data and run it through a decryption routine and then pass the audio to the vocoder portion of the program.
I would recommend a simple key such as 0000 0000 0000 0001 or 1111 1111 1111 1111. Also perhaps a constant 1000hz tone in one part so that a pattern recognition sub routine might be able to speed up confirmation of valid decode.
Maybe a new thread should be started for this so that noone gets sidetracked by posts in the original.
The key loader that I have (T3011DX) will not allow keys like the ones that you describe.
The best that I can do is load 0101 0101 0101 0101 with the Logical ID being 0001.
The best that I can do is load 0101 0101 0101 0101 with the Logical ID being 0001.
Would 7F7F 7F7F 7F7F 7F7F with a Logical ID of ABCD be easier?
You guys tell me and I'll load it and record it. I wish I had more than one alignment cable to pull the audio from more than one radio at a time so I could post the same audio in different formats.
You guys tell me and I'll load it and record it. I wish I had more than one alignment cable to pull the audio from more than one radio at a time so I could post the same audio in different formats.
MattSR
Member
The DES specs mandate odd parity on every 8th bit - thats why your KVL wont accept those keys. This was one of the controversial "features" that NSA recommended back in the 1970s (along with the tickled S-boxes)
Isn't giving out the whole key defeating the point though... As gary alluded to, it would be as simple as merging the P25 DES conformance spec with already available software and its done.
I think giving us the first half of the key would be better - searching 2^28 keys is trivial on modern PC hardware (56 bits is a still bloody expensive and or time consuming exercise!)
Isn't giving out the whole key defeating the point though... As gary alluded to, it would be as simple as merging the P25 DES conformance spec with already available software and its done.
I think giving us the first half of the key would be better - searching 2^28 keys is trivial on modern PC hardware (56 bits is a still bloody expensive and or time consuming exercise!)
MattSR
Member
Dude, key variables are arbitrary - 010101 is just as relevant as a random number such as D45E69 in a well designed encryption algorithm.
This won't help much unless its injected into the radio at a very precise level. Also the IMBE codec standard test pattern is 1011hz - since IMBE is a lossy codec, and makes voiced/unvoiced decisions, tones might not work as expected...
JnglMassiv
Member
We still haven't seen/heard decoded audio with a known key though, right?
MattSR
Member
Dont worry about that - thats the easy part.. The point of the thread is to find an unknown key 
The thing is, I did give out the key to the original contest a few YEARS ago and nobody has sent me decoded audio of the message. One guy got close, but quit due to the time it was taking him. So even with the correct key and type of encryption, it was still extremely difficult.
My original thread was started to find a way to figure out unknown keys. A start was to allow people to decode a known key, which proved to be more difficult than first thought.
My original thread was started to find a way to figure out unknown keys. A start was to allow people to decode a known key, which proved to be more difficult than first thought.
Dont worry about that - thats the easy part.. The point of the thread is to find an unknown key
MattSR
Member
The reasons people got nowhere with the first attempt had more to do with the fact that securenet is proprietary as opposed to the strength of the DES algorithm..
Anyways I just thought it might be a fun exercise
Anyways I just thought it might be a fun exercise
I agree that this threads origninal intent was to decode an unknown message. The idea behind the suggestions I am making is to allow the construction of a decoder packge that will properly decode a known message. Once this routine exists then we can look at various methods to locate an unknown key under various conditions. I dont feel that this would be cheating as in theory anyone with a p25 radio and a decryption module would be starting with this hardware already functional.
- Status
Similar threads
