Files
proxmark3/traces/README.md
T
Matthew CarrollandClaude Opus 5.5 7344c7d5bc lf: fall back to pm3_psk_demod for weak fc/2 PSK when the wave tracker isn't clean
At fc/2 the subcarrier is a one-sample alternation and pskRawDemod_ext finds
wave tops one sample at a time, unfiltered. On a weak tag, sample noise erases
tops; a wave then reads as four samples and counts as a phase shift, and the
bit stream fills with errors although every bit is still in the trace. A T5577
Indala at 22 mm gave the wrong card on 14 of 50 lf search reads.

PSKDemod keeps the tracker's result whenever it is clean, so nothing changes
where master already works. Otherwise, for fc/2, it tries the existing coherent
receiver pm3_psk_demod (now with an optional per-symbol magnitude output) and
keeps that only if it scores well above what level-step leakage can produce,
its bits aren't doubled or constant, and no symbol is weak. The receiver's
output is differential, so one bad decision would flip every later bit: a weak
symbol refuses the trace rather than risk a wrong card (absolute decisions with
weak symbols left to maxErr were tried and gave wrong IDs on dropouts). With no
usable clock from the caller, candidate clocks are tried largest first. Polarity is voted from the
raw level over each phase reversal, the cue pskFindFirstPhaseShift reads at the
first one. Client only; firmware is unchanged.

Base vs this change, same captures:
- T5577 Indala, live lf search: 22 mm 3 right / 14 wrong -> 50 / 0; 24 mm
  0 / 2 -> 50 / 0; 26 mm and 0-20 mm unchanged
- lf t55xx dump, live: 22 and 24 mm blocks 0-2 and page 1 correct 5/5 where
  master reads nothing; unchanged where master reads
- emulated Indala (Chameleon Ultra, clock drifting against the reader):
  lf search right / wrong 56 / 130 -> 91 / 95 of 200; every one of the 95 is
  master's own wrong ID on the same capture
- 27 PSK traces in traces/, weakened with Gaussian noise: correct 341 -> 636 of
  810, wrong IDs 83 -> 21; with recorded Proxmark noise 367 -> 553 of 648,
  33 -> 27. No lost reads and no new wrong IDs in either
- traces/lf_*.pm3 (99): lf search -1 and -1 -c output identical; 1,581
  non-PSK probes: lf search and lf t55xx detect identical
- 200 empty-field captures: no decode, either build

tools/pm3_tests.sh: a weak 24 mm Indala trace must read its ID (fails on
master), and a synthetic NRZ with an fc/2 ripple must not decode as a tag
(the level-step leakage the score check rejects).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-05 14:45:41 -07:00

212 lines
12 KiB
Markdown

# Introduction
* `.pm3` are analog signal files to be loaded with `data load` and displayed with `data plot`
* `.trace` are protocol binary data to be loaded with `trace load` and displayed with `trace list <protocol> 1`
# Analog acquisitions
## LF traces
|filename |description|
|-----------------------------------------|-----------|
|lf_AWID-15-259.pm3 |AWID FSK RF/50 FC: 15 Card: 259 |
|lf_Casi-12ed825c29.pm3 |casi rusco 40 bit (EM410x ID: 12ed825c29)|
|lf_EM4102-1.pm3 |credit card style card EM4102 tag (ID: 010872e77c)|
|lf_EM4102-2.pm3 |credit card style card EM4102 tag (ID: 010872beec)|
|lf_EM4102-3.pm3 |credit card style card EM4102 tag (ID: 010872e14f)|
|lf_EM4102-clamshell.pm3 |Samy's clamshell EM4102 tag (ID: 1f00d9b3a5)|
|lf_EM4102-fob.pm3 |(ID: 0400193cbe)|
|lf_EM4102-thin.pm3 |Samy's thin credit-card style EM4102 tag (ID: 1a0041375d)|
|lf_EM4x05.pm3 |ear tag FDX-B ISO-11784/5 (ID: 6DB0840800F80001 - Application Identifier: 8000, Country Code: 124 (Canada), National ID: 270601654)|
|lf_EM4x50.pm3 |credit card style card EM4x50 tag (ID: DE2A3F00)|
|lf_FDX_Bio-Thermo.pm3 |FDX Animal tag with Biosensor|
|lf_GALLAGHER.pm3 |GALLAGHER tag|
|lf_GProx_36_30_14489.pm3 |G-Prox-II FC: 30 Card: 3949, Format 36b ASK/BIPHASE|
|lf_HID-proxCardII-05512-11432784-1.pm3 |clamshell-style HID ProxCard II card|
|lf_HID-weak-fob-11647.pm3 |HID 32bit Prox Card#: 11647. very weak tag/read but just readable.|
|lf_HomeAgain.pm3 |HomeAgain animal (cat) tag - ID 985121004515220|
|lf_HomeAgain1600.pm3 |HomeAgain animal (cat) tag - ID 985121004515220|
|lf_IDTECK_4944544BAC40E069.pm3 |IDTECK raw 4944544BAC40E069 , PSK, printed "806 082 43084"|
|lf_IDTECK_4944544B351FBE4B.pm3 |IDTECK raw 4944544B351FBE4B , PSK, printed "708 082 14087"|
|lf_IDTECK_idk50_PSK.pm3 |IDTECK (?)|
|lf_Indala-00002-12345678-1A.pm3 |Indala credit-card style card|
|lf_indala_4041x_234_21801.pm3 |Indala 4041X 26-bit|
|lf_Indala-504278295.pm3 |PSK 26 bit indala|
|lf_Indala-a0000000e6bd0e92-weak.pm3 |T5577 as Indala PSK1 RF/32, 24 mm off a Proxmark3 Easy antenna. very weak read (Raw: a0000000e6bd0e92)|
|lf_IOProx-XSF-01-3B-44725.pm3 |IO Prox FSK RF/64 ID in name|
|lf_IOProx-XSF-01-BE-03011.pm3 |IO Prox FSK RF/64 ID in name|
|lf_Keri.pm3 |Keri PSK-3 Key Ring tag (back of tag: 1460 3411)|
|lf_Motorola_0437_00072.pm3 |Motorola Grey clamshell card, old. (RAW: A0000000E308C0C1)|
|lf_NEXWATCH_Nexkey_74755342.pm3 |NEXWATCH, Nexkey ID: 74755342|
|lf_NEXWATCH_Quadrakey-521512301.pm3 |NEXWATCH, Quadrakey ID: 521512301|
|lf_NEXWATCH_Securakey-64169.pm3 |Securakey Tag BitLen: 26, Card ID: 64169, FC: 0x35|
|lf_PAC-8E4C058E.pm3 |PAC/Stanley 20204/21020 PAC8 tag (ID: 8E4C058E)|
|lf_Paradox-96_40426-APJN08.pm3 |PARADOX FC 96 CN 40426|
|lf_TI.pm3 |TI HDX FSK 134.2 / 123.2kHz, zerocross line acquisition at 2 MHz|
|lf_Transit999-best.pm3 |Transit 999 format (UID 99531670)|
|lf_VeriChip_1022000000084146.pm3 |VeriChip, epoxy encased glasschip (ID: 1022-00000000084146) |
|lf_VISA2000.pm3 |VISA2000 ASK/MAN RF/64, Card: 480518|
## LF test traces
|filename|description|
|--------|-----------|
|lf_Q5_mod-* |Q5 configured to emit `00 01 02 03 04 05 06 07 08 09 0A 0B` under various modulation schemes|
|lf_ATA5577_* |ATA5577 configured to emulate various techs as suggested in the Proxmark3 clone commands|
|lf_ATA5577.txt |Description on how lf_ATA5577_* were generated|
|lf_NRZ_RF64_fc2-ripple_synthetic.pm3 |synthetic, no tag: random NRZ RF/64 at +-40 with first-order settling, a +-1.5 fc/2 alternation and 0.3 rms noise. must not decode as PSK|
## LF sniffed traces
|filename|description|
|--------|-----------|
|lf_sniff_blue_cloner_em4100.pm3 |Sniffing of blue cloner writing an EM4100 on T5577 and EM4305|
|lf_sniff_ht2-BC3B8810-acg-reader.pm3 |Sniffing of Hitag2 being read by an HID ACG LF Multitag reader|
|lf_sniff_ht2-BC3B8810-frosch-reader.pm3 |Sniffing of Hitag2 being read by a Frosch Hitag reader|
|lf_sniff_ht2-BC3B8810-rfidler-reader.pm3 |Sniffing of Hitag2 being read by a RFIDler|
|lf_ht2_pwd_sniff.trace |Sniffing of a genuine Hitag2 in password mode being read by a Paxton reader|
|lf_ht2_crypto_sniff.trace |Sniffing of a Hitag2 in crypto mode, showing the NrAr exchange|
|lf_ht2_pwd_sim.trace |`lf hitag sim` answering a Paxton reader in password mode, recorded by the simulating Proxmark|
## HF traces
|filename|description|
|--------|-----------|
|hf_14b_raw_050008_resp.pm3 |Response of 14b card to `hf 14b raw -c 050008`|
|hf_14b_raw_0600_st_sri512.pm3 |Response of ST SRI512 to `hf 14b raw -c 0600`|
|hf_14b_raw_0600_st_sri512_collision.pm3 |Same but with two cards, showing the collisions in answers|
|hf_14b_raw_10_ask_ctx.pm3 |Response of ASK CTx to `hf 14b raw -c 10`|
|hf_14b_raw_010fxxxxxxxx_innovatron.pm3 |Response of 14b' card to `hf 14b raw -c -k 010fxxxxxxxx`|
## HF sniffed traces
|filename|description|
|--------|-----------|
|hf_sniff_14b_scl3711.pm3 |`hf sniff 15000 2` <> `nfc-list -t 8`: PUPI: c12c8b1b AppData: 00000000 ProtInfo: 917171|
# Demodulated acquisitions
## HF demodulated traces
|filename|description|
|--------|-----------|
|hf_14a_reader_4b.trace |Execution of `hf 14a reader` against a 4b UID card|
|hf_14a_reader_4b_rats.trace |Execution of `hf 14a reader` against a 4b UID card with RATS|
|hf_14a_reader_7b_rats.trace |Execution of `hf 14a reader` against a 7b UID card with RATS|
|hf_14a_mfu.trace |Reading of a password-protected MFU|
|hf_14a_mfuc.trace |Reading of a UL-C with 3DES authentication|
|hf_14a_mfu-sim.trace |Trace seen from a Proxmark3 simulating a MFU|
|hf_14b_reader.trace |Execution of `hf 14b reader` against a card|
|hf_14b_cryptorf_select.trace |Sniff of libnfc select / anticollision ofa cryptoRF tag|
|hf_14b_xerox_info.trace |Execution of `hf xerox info` against a tag|
|hf_15_reader.trace |Execution of `hf 15 reader` against a card|
|hf_mfp_mad_sl3.trace |`hf mfp mad`|
|hf_mfp_read_sc0_sl3.trace |`hf mfp rdsc --sn 0 -k ...`|
|hf_visa_apple_ecp.trace |Sniff of VISA Apple ECP transaction|
|hf_visa_apple_normal.trace |Sniff of VISA Apple normal transaction|
|hf_visa_apple_transit_bypass.trace |Sniff of VISA Apple transaction bypass|
|hf_mfdes_sniff.trace |Sniff of HID reader reading a MIFARE DESFire SIO card|
|hf_iclass_sniff.trace |Sniff of HID reader reading a Picopass 2k card|
|hf_mf_hid_sio_sim.trace |Simulation of a HID SIO MFC 1K card|
|hf_14a_mful_aes_defaultkey.trace |Reading of a UL-AES with AES authentication|
|hf_14a_mful_aes_defaultkey_secure_message.trace|Reading of a UL-AES with AES authentication, GetVersion with secure message|
## LF demodulated traces
|filename|description|
|--------|-----------|
|lf_hitag_crypto_dump.trace |Execution of `lf hitag dump --crypto` against Hitag2 card in crypto mode|
## LF emulator dumps
Tag images for `lf hitag eload`, used to simulate a Hitag 2 in each of its
operating modes. Byte 0 of page 3 is the configuration byte that selects the
mode, so the files differ mainly in that one byte.
|filename|description|
|--------|-----------|
|lf-hitag-CE129911-dump.bin |Hitag2 in password mode (config `0x06`), password `BDF5E846`|
|lf_ht2_crypto.bin |Hitag2 in crypto mode (config `0x0E`), key `4F4EBDF5E846`|
|lf_ht2_crypto_1CAEFD13.bin |Hitag2 in crypto mode (config `0x0E`)|
|lf_ht2_publicA.bin |Hitag2 in Public Mode A (config `0x02`), Manchester, pages 4-5, carries an EM4102 payload|
|lf_ht2_publicB.bin |Hitag2 in Public Mode B (config `0x00`), biphase 32 T0, pages 4-7|
|lf_ht2_publicC.bin |Hitag2 in Public Mode C (config `0x04`), biphase 64 T0, pages 4-7|
## MIFARE DESFire card images
Card images for `hf mfdes eload`, used by `hf mfdes sim`. All three are in the
`mfdes v1` dump format documented in `doc/mfdes_dump_format.md`.
The 8K images come off a genuine MIFARE DESFire EV1 8K, UID `04268512A25680`,
batch `B9 0C 17 49 70`, week 27 / 2017, hardware `04010101001A05`, software
`04010101041A05`.
|filename|description|
|--------|-----------|
|hf-mfdes-ev1-8k-empty.json |EV1 8K straight after `formatpicc`. PICC only, default 2TDEA key, 7936 bytes free|
|hf-mfdes-ev1-4k-empty.json |The same image edited to present as an EV1 4K. **Derived, not dumped from 4K silicon** -- the storage size byte of `VersionHW`/`VersionSW` is changed from `0x1A` to `0x18`, and `FreeMem` set to 3840, the nominal 4096 minus the 256 bytes of overhead the 8K part shows|
|hf-mfdes-ev1-8k-test.json |The same card with three applications and all five EV1 file types, for exercising a simulation|
|hf-mfdes-ev1-8k-test-keys.json |Application keys for the above, in the `mfdes v2` format `hf mfdes chk -j` writes|
|hf-mfdes-ev1-8k-empty-keys.json|Key file for the empty 8K image. Empty by construction, a blank card has no applications|
|hf-mfdes-ev1-4k-empty-keys.json|Key file for the empty 4K image. Likewise empty|
### Keys
Every key on this card is the counting sequence `01 02 03 .. 10`, extended to
`.. 18` for the 24 byte 3TDEA keys. The PICC master key is 2TDEA
`0102030405060708090A0B0C0D0E0F10`; each application uses its own algorithm,
and all of its keys are that algorithm's length of the same sequence.
Deliberately not the all-zero key, and not a key whose two halves match. A
16 byte DES/2TDEA key string whose second half equals the first is handled as a
single DES key by the PICC -- during authentication *and* session key
generation (M134034 8.1) -- and the all-zero key is the common case of that. A
card image keyed with it exercises only the degenerate path: an implementation
that derives the wrong session key still authenticates, and only fails later on
the first MACed or enciphered frame. This sequence has distinct halves, so a
wrong derivation shows up immediately.
The `mfdes v2` key files carry **application** keys only -- that format has no
place for the PICC master key, because `hf mfdes chk` only walks the application
list and takes the PICC key from `-k`. Hand it in explicitly:
```
hf mfdes dump -n 0 -t 2TDEA -k 0102030405060708090A0B0C0D0E0F10 --keys traces/mifare/hf-mfdes-ev1-8k-test-keys.json
```
Note that every key is also stored inside the dump itself, under each
application's `Keys` object, so `eload` and `sim` need no key file at all --
the key files are for reading or rebuilding the physical card.
`hf-mfdes-ev1-8k-test.json` holds:
|AID|keys|ISO DF name|files|
|---|----|-----------|-----|
|`010203`|AES, 3 |`test1`|`01` standard 256 B free access, `02` backup 128 B MAC mode, rights `1200`|
|`112233`|2TDEA, 2 |`test2`|`00` value, 1000 with limits [0..10000], `01` linear record, 2 of 8 records, 16 B each|
|`AABBCC`|3TDEA, 5 |`test3`|`05` cyclic record, 1 of 4 records, 24 B each, `03` standard 64 B free access|
```
hf mfdes eload -f traces/mifare/hf-mfdes-ev1-8k-test.json
hf mfdes sim
```
## ST25TB card images
`14b v2` dump format, 128 EEPROM blocks of four bytes followed by the system
block. The file name has to carry the UID, `hf 14b view` reads it from there to
work out the chip type and, for MyKey, to derive the session key.
|filename|description|
|--------|-----------|
|hf-14b-D0021F673CB26556-dump.json |ST25TB04K carrying a COGES v3 MyKey, reset state: no vendor bound, zero credit, empty transaction ring. Blocks 7 and 8 are write protected|
```
hf 14b view -f traces/hf-14b-D0021F673CB26556-dump.json
```
This is the vector behind `hf 14b view --selftest`, which checks the block
scrambler, the session key and all 48 block checksums against it.