proxmark3/tools/hitag2crack

HiTag2 Cracking Suite

Authors:

• Attacks 1, 2, 3, 4 : Kevin Sheldrake kev@headhacking.com
• Attacks 5, 5gpu : anonymous, based on https://github.com/factoritbv/hitag2hell by FactorIT B.V.
• Attack 5opencl : Gabriele Gristina gabriele.gristina@gmail.com, based on 5gpu

Introduction

These tools are implementations of the four attacks detailed in the papers, Gone In 360 Seconds - Hijacking With HiTag 2 by Roel Verdult, Flavio Garcia and Josep Balasch, and Lock It And Still Lose It - on the (In)Security of Automotive Remote Keyless Entry Systems by Flavio Garcia, David Oswald, Timo Kasper and Pierre Pavlides. The first three attacks come from the first paper and the fourth attack comes from the second paper.

note There are no Proxmark3 client implemented commands for these attacks. Only separate executables to be compiled and run on your own system. No guarantees of working binaries on all systems. Some work on linux only. There is no easy way to extract the needed data from a live system and use with these tools. You can use the RFIdler device but the Proxmark3 client needs some more love. Feel free to contribute.

Attack 1

Attack 1 is a nonce replay and length extension attack. This is an attack on a single HiTag2 RFID tag, given a single encrypted nonce and challenge response value pair (nR, aR) for the tag's UID. The attack runs entirely on the Proxmark3 with it acting like a RWD that replays the same encrypted nonce and challenge response pair for every interaction; this fixes the key stream that the tag's PRNG outputs to the same stream for every interaction.

By brute forcing a subset of the encrypted command space, the Proxmark3 finds a single valid encrypted command - invalid commands return a known unencrypted error response so finding a valid one is simply a case of trying different values until a response other than the error response is received.

It then bit flips the valid encrypted command to find the other 15 valid encrypted commands. By knowing the contents of page 0 - it's the UID that is presented in clear at the start of each interaction - it tries each encrypted response in turn, assuming each to be the encrypted version of 'read page 0 non-inverted' and each response to be the encrypted version of page 0.

For each attempted command, it calculates the key stream that would have correctly generated the encrypted command and response: command ++ response XOR key stream = encrypted command ++ encrypted response therefore: key stream = command ++ response XOR encrypted command ++ encrypted response

It then tests the potentially recovered key stream by creating an encrypted command that consumes as much of it as possible, re-initialising with the same encrypted nonce and challenge response pair (to set the key stream to the same stream as that which produced the encrypted command response it is testing), and then sending this extended encrypted command. If the response is not the error response, then the key stream is valid and the response is the encryption of the page 0 contents (the UID).

When one of the valid encrypted commands satisfies this situation, the recovered key stream must be the output of the PRNG for the given encrypted nonce and challenge response pair.

The Proxmark3 then uses this key stream to encrypt commands and decrypt the responses, and therefore requests the contents of all 8 pages. Pages 1 and 2 contain the encryption key.

Attack 2

Attack 2 is a time/space trade off to recover the key for situations where the tag has been configured to prevent reading of pages 1 and 2. This attack uses a pre-computed table of 2^37 PRNG states and resultant PRNG output, sorted on the PRNG output. The Proxmark3 is used to recover 2048 bits of key stream using a modification of attack 1 and this is used to search the table for matching PRNG output. When the output is found, it is tested for validity (by testing previous or following PRNG output) and then the PRNG state is rolled back to the initialisation state, from which the unencrypted nonce and key can be recovered.

Attack 3

Attack 3 is a cryptanalytic attack that focuses on the RWD and a bias in the PRNG output. By capturing 136 encrypted nonce and challenge response pairs, candidates for the first 34 bits of the key can be identified, and for each the remaining 14 bits can be brute forced.

Attack 4

Attack 4 is a fast correlative attack on the key based on a number of captured encrypted nonce and challenge response pairs (up to 32, but 16 usually sufficient). It starts by guessing the first 16 bits of the key and scores all these guesses against how likely they are to be the correct key, given the encrypted nonces and the keystream they should produce. Each guess is then expanded by 1 bit and the process iterates, with only the best guesses taken forward to the next iteration.

Attack 5

Attack 5 is heavily based on the HiTag2 Hell CPU implementation from https://github.com/factoritbv/hitag2hell by FactorIT B.V., with the following changes:

• Main takes a UID and 2 {nR},{aR} pairs as arguments and searches for states producing the first aR sample, reconstructs the corresponding key candidates and tests them against the second nR,aR pair;
• Reuses the Hitag helping functions of the other attacks.

Attack 5gpu

Attack 5gpu is identical to attack 5, simply the code has been ported to OpenCL to run on GPUs and is therefore much faster than attack 5.

Attack 5opencl

Attack 5opencl is an optimized OpenCL version based on 5gpu. It runs on multi GPUs/CPUs and is faster than 5gpu.

Usage details: Attack 1

Attack 1 requires a valid tag and a valid encrypted nonce and challenge response pair. The attacker needs to obtain a valid tag and then use this to obtain a valid encrypted nonce and challenge response pair. This can be achieved by using the Proxmark3 lf hitag sniff command, placing the coil on the RWD and presenting the valid tag. The encrypted nonce and challenge response pairs can then be read out.

note the Proxmark3 hitag sniff command isn't good enough yet to collect the needed data.

TODO example

Usage details: Attack 2

Attack 2 requires the same resources as attack 1, plus a pre-computed table. The table can be generated on a disk with >1.5TB of storage, although it takes some time (allow a couple of days, privilege SSD). This can be achieved by using the Proxmark3 lf hitag sniff command, placing the coil on the RWD and presenting the valid tag. The encrypted nonce and challenge response pairs can then be read out.

note the Proxmark3 hitag sniff command isn't good enough yet to collect the needed data.

TODO example

Usage details: Attack 3

Attack 3 requires only interaction with the RWD and does not require a valid tag, although it does require a HiTag2 tag that the RWD will initially respond to; e.g. you could potentially use any HiTag2 tag as long as the RWD starts the crypto handshake with it. It requires >=136 encrypted nonce and challenge response pairs for the same tag UID.

note the Proxmark3 hitag sniff command isn't good enough yet to collect the needed data.

TODO will be ht2 sim or sniff with actual tag ?

Usage details: Attack 4

Attack 4 requires the same information as attack 3, but only 16-32 encrypted nonce and challenge response pairs are required.

note the Proxmark3 hitag sniff command isn't good enough yet to collect the needed data.

TODO example

Usage details: Attack 5

Attack 5 requires two encrypted nonce and challenge response value pairs (nR, aR) for the tag's UID.

TODO example

Usage details: Attack 5gpu/5opencl

Attacks 5gpu and 5opencl require two encrypted nonce and challenge response value pairs (nR, aR) for the tag's UID.

TODO example

5opencl supports a number of additional parameters, see crack5opencl/README.md for details.

Usage details: Next steps

Once the key has been recovered using one of these attacks, the Proxmark3 can be configured to operate as a RWD and will capture tags using that key.

TODO example

Tags can be copied with standard Proxmark3 commands.

TODO example