P25 cryptography security video - RUXCON 2011

[MattSR]

Hi All,

This is a video of a presentation I recently did at a security conference in Melbourne, Australia. In the audience were various Federal and State police agencies, as well as the Aussie equivalent of the NSA (we call them the DSD)

Ruxcon 2011: APCO P25 Security Revisited: The Practical Attacks - YouTube

It includes a demo of DES-OFB decryption done in entirely in GNUradio on a PC (yes the voice traffic is recovered and played through the PA system)

Everything in the presentation is correct and verifiable from the references in the security paper we published earlier this year.

Please note - I don't want this thread to degrade into a "crypto can't be broken" war. The point of this video is to educate and inform people about how things work in reality. The silence known plaintext attack works and is verified in the video as working with DES-OFB.

Regards,
Matt

 

[MattSR]

Fast forward to 38 mins if you just want to skip to the software defined P25 decryption demo.

 

[nycap]

Hi All,

This is a video of a presentation I recently did at a security conference in Melbourne, Australia. In the audience were various Federal and State police agencies, as well as the Aussie equivalent of the NSA (we call them the DSD)

Ruxcon 2011: APCO P25 Security Revisited: The Practical Attacks - YouTube

It includes a demo of DES-OFB decryption done in entirely in GNUradio on a PC (yes the voice traffic is recovered and played through the PA system)

Everything in the presentation is correct and verifiable from the references in the security paper we published earlier this year.

Please note - I don't want this thread to degrade into a "crypto can't be broken" war. The point of this video is to educate and inform people about how things work in reality. The silence known plaintext attack works and is verified in the video as working with DES-OFB.

Regards,
Matt

but what gear do you use to compare the plain text to the cypher text and how long does it take to retreive the key?

 

[MattSR]

Any of these - Pico Computing - the FPGA Computing Experts

The SC5 does a full DES search in 2.41 days. It tests 345 Billion DES keys per second.

 

[mancow]

That's wild. 345 billion keys!

What about ADP, has anyone messed with that?

 

[MattSR]

Watch the presentation at 21:24 mins. The ADP algorithm is explained and the ley recovery done using Nvidia GPUs.

 

[nycap]

my hats off to you sir. i am assuming this will crack RC within hours? did you program the FPGAs or is there a ready made key recovery program available?

 

[jets1961]

Could someone jog my memory, what is the cypher strength of DES-OFB encryption?

I will review the complete video tonight but I have to ask was it done in real time. So was the key unknown and deciphered in real time? I did look at that part of the video and that is the way it appears.

Thanks, Joe

 

[nycap]

it takes a couple days to get the key. then you put the key in the radio and listen in real time.

 

[nycap]

Any of these - Pico Computing - the FPGA Computing Experts

The SC5 does a full DES search in 2.41 days. It tests 345 Billion DES keys per second.

do you know what this machine costs?

 

[RadioFlyer2012]

Guessing is one way to do it lol
I assume they retail them for 15k, will inquire.

 

[RadioFlyer2012]

Forum should add ability to edit previous posts.
Price 100-200k, and considering only 56des, and the fact that codes are changed daily or multiple times a day, save your money.

 

[pepsima1]

Hi All,

This is a video of a presentation I recently did at a security conference in Melbourne, Australia. In the audience were various Federal and State police agencies, as well as the Aussie equivalent of the NSA (we call them the DSD)

Ruxcon 2011: APCO P25 Security Revisited: The Practical Attacks - YouTube

It includes a demo of DES-OFB decryption done in entirely in GNUradio on a PC (yes the voice traffic is recovered and played through the PA system)

Everything in the presentation is correct and verifiable from the references in the security paper we published earlier this year.

Please note - I don't want this thread to degrade into a "crypto can't be broken" war. The point of this video is to educate and inform people about how things work in reality. The silence known plaintext attack works and is verified in the video as working with DES-OFB.

Regards,
Matt

Matt: I was just wondering if anything more has come to light about this technology since this was released a couple of years ago for the scanning community on encryption?? This is awesome technology

 

[AZScanner]

Yep, well when they break AES-256 let me know, because that's what our local P25 system uses. Neat concept though.

-AZ

 

[balibago]

A lot of these agencies have gone ADP which at the very least we could monitor covertly. Time to give us good guys a break.

 

[brandon]

Matt: I was just wondering if anything more has come to light about this technology since this was released a couple of years ago for the scanning community on encryption?? This is awesome technology

Here is a good website to check out
OP25

World's cheapest P25 receiver w/ decryption: GNU Radio + OP25 + $20 RTL2832 DVB-T Dongle - YouTube

Good luck getting the keys

 

[MattSR]

That's the exact website for the exact project that the video and thread is about, and it's my yellow XTS5000 in that video too.

Getting the keys recovered is the whole point of the OP25 project....

 

[balibago]

Is there any way an ADP for dummies download or book will ever come out. It looks like you have to be sort of a math or computer wiz to do this. Our big problem here in the states is these agencies using ADP to "cover their asses". There is no real need to keep the 2 percent or so Americans who scan from listening in.

 

[pepsima1]

Any of these - Pico Computing - the FPGA Computing Experts

The SC5 does a full DES search in 2.41 days. It tests 345 Billion DES keys per second.

Any comments on this site and the super computer here.

COPACOBANA - Special-Purpose Hardware for Code-Breaking

 

[balibago]

You don't need a super computer

This ADP encryption doesn't require a super computer to break. However what is needed is some type of black hat software which can emulate the logic in the cipher as well as software to do the search of the 1 Trillion (not so big a number nowadays) or so keys in it and check them against the standard moment of silence which all P25 radios have at the end of their transmissions. It amazes me to no end how some hacker has not stepped up to the plate and put such a program out there in the wild so to speak.