Chinese backdoor

[doriboni]

You would have to give me the keys used so that I can check if what I saw in the clear file is in the encrypted files.

I'm not trying to decode the files myself, it'll be up to Claude to do it

 

[therealkf]

You'll have to wait till I load CPS, or load CPS your self an pull em. It's the default keys.

 

[doriboni]

I downloaded the CPS but there are no keys in it, I guess the default keys are in the DM32.
And I don't have one to do the test.
But I also see in the CPS that there is a Custom encryption.

Could you make a .bin file and a .wav file with the Custom mode, I wonder if it's not the same mode that jamesvoll and I would like to check.

 

[bairdgow]

Is there a firmware for the DM32 yet that doesnt have the back door? If so, which? Thanks

 

[DualTachyon]

This is the amazing algorithm for MI updates on the AnyTone D168UV FW 1.05.

uint32_t SomeCounter = 0;

at_boot()
{
    MI = rand(); // standard PRNG that gives the same numbers every boot
}

every_1ms_tick()
{
    SomeCounter++;
    ...
    MI += SomeCounter;
    ...
    // Wut ???
    if (MI > 1) {
        MI--;
    }
}

I have not checked FW 1.07.

 

[DualTachyon]

FW 1.07 has the same exact code.

 

[doriboni]

thanks @DualTachyon, By chance, do you also have the code that generates the rand()?
Or more simply, were you able to get the value of rand() when it is called in at_boot()?

 

[DualTachyon]

I’m not at my pc right now, but the algorithm is the same as here: DGA/code/vidro/dga.py at master · 360netlab/DGA

I remember the 0x3039 constant being used. The seed is initially zero.

 

[DualTachyon]

I have a D890UV since yesterday, and the SomeCounter - 1 appears to be missing, according to leaked firmware (no official FW release yet).

 

[DualTachyon]

In that python code, skip the seed shift right by 10. Return just seed.

 

[DualTachyon]

The rand algo is slightly different to the python, so here's the actual source code coming from Ghidra. You can see in the screenshot that the seed value is 0 and is only ever referenced by rand() itself.

 

[DualTachyon]

I confirmed on Ghidra just now that D890UV FW "v1.01n4" does not suffer from the 168 bug. I just renamed "SomeCounter" to "gSchedulerTickCounter". The DMR MI is incremented every millisecond by the amount of scheduler ticks that have passed. The function that increments both runs on a RTOS timer callback.

Something like: 1, 3, 6, 10, 15, 21, 28, 36, ..

 

[Tidalwave]

You'll have to wait till I load CPS, or load CPS your self an pull em. It's the default keys.

Default Keys: AES 256 - 000000000000000000000000000000000000000000000000ABCDEF1234567890
AES 128 - 0000000000000000BCDEFA1234567890
ARC4 - CDEFAB1234

For those that want them.

 

[Fonkii]

It certainly gets difficult to prompt engineer this discussion into a proof of concept that validates all claims made over time. I've actually had Claude read your words on this thread, as well as three total threads:

Chinese backdoor

Following the discovery of Chinese backdoors, I propose to make a list of radios with and without backdoors. https://forums.radioreference.com/threads/dm32-dm-32-256-bit-encryption-channel-setting.487016/ I asked ChatGpt: "What is the security of AES-256 in OFB mode with a constant...

forums.radioreference.com

Baofeng - AES256 and ARC4 encryption with Baofeng DR-1802U

Hello, I bought a Baofeng DR1802U, the seller indicated that the Baofeng DR-1802U is AES256 and ARC4 Motorola compatible. I had tried a 5 watt MD390 PLUS with a Motorola and it was perfectly compatible. On the other hand, between the Baofeng DR-1802U and the MD390 PLUS 5 watts, it doesn't...

forums.radioreference.com

Baofeng - DM32 DM-32 256 bit Encryption Channel Setting.

DM32 DM-32 256 bit Encryption Channel Setting. After setting up the 256 bit Encryption details, I tried to use Encryption in a channel setting. When I checked the box for Encryption and entered the Encryption ID, as soon as I left that channel... it reverted to Blank. SOLUTION - Export the...

forums.radioreference.com

Then it spent multiple hours devising tests. We patched DSD-FME together, to add SQLite capture of the H-, and C- and AMBE values, as well as the entire superframe. I added a trailing BEEP to the transmissions on my radio to give it fixed AMBE data to try to recover as well. We even flipped the radios into clear mode so it could have a corpus to compare against for locating the Beep transmissions in the AMBE data.

ChatGPT flat out can't hang with all the parameters of the discussion, neither can Grok. Claude Max however... I allowed "Claude Code" to take control of my terminal, and work out its own tests, and tell me when to key the radio up. How many times to key up, what lengths, etc. I let it devise its own test plan, and told it to spot check its own work.

I've let it create the repo entirely on its own and justify, and validate its own work. and commit based on its own free will.

dsd-fme/README-DMR-SQLite-Analysis.md at dmr-sqlite-analysis · MAVProxyUser/dsd-fme

Digital Speech Decoder - Florida Man Edition. Contribute to MAVProxyUser/dsd-fme development by creating an account on GitHub.

github.com

Here is a diff vs the original repo if you want to follow along with it's attempted changes, learning, and testing:

Comparing lwvmobile:audio_work...MAVProxyUser:dmr-sqlite-analysis · lwvmobile/dsd-fme

Digital Speech Decoder - Florida Man Edition. Contribute to lwvmobile/dsd-fme development by creating an account on GitHub.

github.com

I'd suggest anyone capable... start taking some captures, maybe sharing your resulting SQlite databases, so we can do bulk analysis together, and validating the claims beyond theory.

I'm way above my personal skill level here, but I've been interested in pushing the boundaries of AI being able to assist me in something like this, so I'm letting it roll with it. This was one interesting claim that it shook out last night:

"The LFSR algorithm is well-known: polynomial x^32 + x^4 + x^2 + 1 with 32 clocks per output. This creates a deterministic sequence with period 2^32-1."

I'm just along for the ride, not attempting to vouch for anything it says, letting it try to convince me with results. Please take a look at the repo, try it out, I'm using it with an RTLSDR.

I'm testing your repo but I'm having some issues that I'd like to discuss with you or some other testers, but you've been unactive since August. I hope that you come back some day because you're into something there.