AYCWABTU

overview

AYCWABTU is a proof of concept for a brute force control word calculation tool for the common scrambling algorithm used in digital video broadcasting.

AYCWABTU is not useful for live decryption of pay TV channels because the search for one key needs much more time than the key renewal interval. Majority of channels change keys multiple times a minute and AYCWABTU needs months to brute force one key. AYCWABTU is intended as proof of concept, and is not intended to be used for illegal purposes. The author does not accept responsibility for ANY damage incurred by the use of it.

It uses parallel bit slice technique. Other csa parallel bit slice implementations (like libdvbcsa) are meant for stream processing. They encrypt or decrypt many packets with one key. AYCWABTU uses parallel bit slice for decrypting one packet with many keys.

features

• fast brute force key calculation due to bit sliced crack algorithm (sse2 and 32 bit versions available)
• open source. License: GPL
• read three encrypted data packets from ts file with many checks for valid data
• writes a small probe ts file with these packets for sharing and distributed attack
• test frame included to make sure, it really finds the keys. Also suitable for other brute force tools
• written in C. Developed in Visual Studio 2013, tested with gcc 4.8.2+cygwin 2.844
• much potential for speed improvements

to do list

• speed improvement, see ideas below
• multi threading
• support for 256 bits parallel with advanced vector extensions AVX
• OpenCL support
• optimize the block sbox boolean equations. Only slightly faster with 128 bits. See da_diett.pdf Chpt. 3.1
• Ctrl-C handling on linux/windows

speed optimization ideas

• most important: OpenCL support (not only CUDA, please!).
• check, why aycw_block_sbox(&sbox_out) fails in gcc, possible speedup ~19%
• block decrypt first (does not depend on stream). Then stream afterwards, stop XORing immediately if foreseeable there is no PES header

developers

• after changing the code, run tests with SELFTEST enabled to make sure the algorithm still works. It's too easy to break things.
• run "make check"
• test all the batch size implementations
• share your benchmark values in the pull request
• publish all your work please, AYCWABTU is released under GPL

credits

• FFdecsa, Copyright 2003-2004, fatih89r
• libdvbcsa, http://www.videolan.org/developers/libdvbcsa.html
• ANALYSIS OF THE DVB COMMON SCRAMBLING ALGORITHM, Ralf-Philipp Weinmann and Kai Wirt, Technical University of Darmstadt Department of Computer Science Darmstadt, Germany
• On the Security of Digital Video Broadcast Encryption, Markus Diett
• http://en.wikipedia.org/wiki/Common_Scrambling_Algorithm
• Breaking DVB-CSA, Erik Tews, Julian Waelde, Michael Weiner, Technische Universitaet Darmstadt
• TSDEC - the DVB transport stream offline decrypter, http://sourceforge.net/projects/tsdec/
• http://csa.irde.to This page disappeared unfortunately but it still accessible here https://web.archive.org/web/20040903151642/http://csa.irde.to/
• and last not least my good friend johnf

"Sorry, it is hard to understand and modify but it was harder to design and implement!!!" fatih89r

Have fun.

ganymede