Datasets:

lkk688
/

rf-stream-otfs

Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
run_id: string
modulation: string
bps: int64
fc_MHz: double
fs_MHz: double
rx_gain_dB: double
gain_sweep: list<item: null>
  child 0, item: null
ref_seed: int64
ref_len: int64
total_cap: int64
good_pkts: int64
decode_rate: double
ber_per_gain: struct<>
gate_model: string
gate_threshold: double
end_time: string
splits: list<item: struct<split: string, src: string, n_files: int64, size_bytes: int64, description: string (... 38 chars omitted)
  child 0, item: struct<split: string, src: string, n_files: int64, size_bytes: int64, description: string, key_resul (... 26 chars omitted)
      child 0, split: string
      child 1, src: string
      child 2, n_files: int64
      child 3, size_bytes: int64
      child 4, description: string
      child 5, key_result: string
      child 6, status: string
staged_at: string
tx_iq_files: list<item: string>
  child 0, item: string
to
{'staged_at': Value('string'), 'splits': List({'split': Value('string'), 'src': Value('string'), 'n_files': Value('int64'), 'size_bytes': Value('int64'), 'description': Value('string'), 'key_result': Value('string'), 'status': Value('string')}), 'tx_iq_files': List(Value('string'))}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 99, in get_rows_or_raise
                  return get_rows(
                         ^^^^^^^^^
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                         ^^^^^^^^^^^^^^^^^^^^^
                File "/src/services/worker/src/worker/utils.py", line 77, in get_rows
                  rows_plus_one = list(itertools.islice(ds, rows_max_number + 1))
                                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2690, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2227, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 2251, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 494, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/iterable_dataset.py", line 384, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 299, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/packaged_modules/json/json.py", line 128, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                             ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2321, in table_cast
                  return cast_table_to_schema(table, schema)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.12/site-packages/datasets/table.py", line 2249, in cast_table_to_schema
                  raise CastError(
              datasets.table.CastError: Couldn't cast
              run_id: string
              modulation: string
              bps: int64
              fc_MHz: double
              fs_MHz: double
              rx_gain_dB: double
              gain_sweep: list<item: null>
                child 0, item: null
              ref_seed: int64
              ref_len: int64
              total_cap: int64
              good_pkts: int64
              decode_rate: double
              ber_per_gain: struct<>
              gate_model: string
              gate_threshold: double
              end_time: string
              splits: list<item: struct<split: string, src: string, n_files: int64, size_bytes: int64, description: string (... 38 chars omitted)
                child 0, item: struct<split: string, src: string, n_files: int64, size_bytes: int64, description: string, key_resul (... 26 chars omitted)
                    child 0, split: string
                    child 1, src: string
                    child 2, n_files: int64
                    child 3, size_bytes: int64
                    child 4, description: string
                    child 5, key_result: string
                    child 6, status: string
              staged_at: string
              tx_iq_files: list<item: string>
                child 0, item: string
              to
              {'staged_at': Value('string'), 'splits': List({'split': Value('string'), 'src': Value('string'), 'n_files': Value('int64'), 'size_bytes': Value('int64'), 'description': Value('string'), 'key_result': Value('string'), 'status': Value('string')}), 'tx_iq_files': List(Value('string'))}
              because column names don't match

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

RF-Stream-PHY OTFS — Raw IQ and Decode Captures

Companion dataset for the paper:

Hardware-Validated OTFS+LDPC on PlutoSDR: Synthetic Doppler Injection and Monostatic-Loopback Mobility Sensing, K. Liu, submitted IEEE Globecom 2026.

Code: https://github.com/lkk688/RF-Stream-PHY (subproject OTFS/).

Total: 12 experiment runs, 2.95 GB, captured 2026-05-31.

Splits

Split	Description	Key result	# files	Size (MB)
`closed_grid_0db_otfs/`	OTFS+LDPC OTA, fc=2.45 GHz, tx_gain=0 dB, 500 caps, paper Sec. 5.2	179/500 = 35.8% pkt rate	502	83.4
`closed_grid_0db_ofdm/`	OFDM+LDPC OTA, fc=2.45 GHz, tx_gain=0 dB, step10 RX with NPZ save; LDPC post-decoded by ofdm_ldpc_postdecode.py, paper Sec. 5.2	161/489 attempts = 32.9% pkt rate (post-LDPC)	492	1112.4
`closed_grid_n5db_otfs_v2/`	OTFS+LDPC OTA, fc=2.45 GHz, tx_gain=-5 dB, 500 caps, after deterministic Pluto gain reset (3x gain rewrite); channel below LDPC cliff so 0/500	0/500 = 0.0% (snr_post median ~-4 dB)	502	82.7
`closed_grid_n5db_ofdm_v2/`	OFDM+LDPC OTA, fc=2.45 GHz, tx_gain=-5 dB, step10 RX, back-to-back with OTFS-n5db_v2; channel past LDPC cliff	0/92 attempts = 0%, raw BER median 0.47	205	446.1
`ota_doppler/otfs_fd0/`	OTFS+LDPC, fc=2.42 GHz, synthetic Doppler fd=0 Hz (multipath only)	71/500 = 14.2%	502	90.4
`ota_doppler/otfs_fd1000/`	OTFS+LDPC, fc=2.42 GHz, synthetic Doppler fd=1000 Hz	19/500 = 3.8%	502	88.8
`ota_doppler/otfs_fd3000/`	OTFS+LDPC, fc=2.42 GHz, synthetic Doppler fd=3000 Hz (OTFS k_max ceiling)	0/500 = 0.0%	502	84.5
`ota_doppler/ofdm_fd0/`	OFDM+LDPC, fc=2.42 GHz, synthetic Doppler fd=0 Hz	0/23 attempts = 0% (raw BER 0.47)	107	232.4
`ota_doppler/ofdm_fd1000/`	OFDM+LDPC, fc=2.42 GHz, synthetic Doppler fd=1000 Hz	0/46 attempts = 0%	172	392.5
`ota_doppler/ofdm_fd3000/`	OFDM+LDPC, fc=2.42 GHz, synthetic Doppler fd=3000 Hz	0/38 attempts = 0%	112	246.0
`monostatic/static/`	Single-Pluto monostatic loopback, OTFS+LDPC, RX antenna cable-extended and HELD STILL, paper Sec. 5.4	0/300 = 0%, mean CFO -8 Hz (vs -4549 Hz two-Pluto)	302	46.4
`monostatic/motion/`	Single-Pluto monostatic loopback, OTFS+LDPC, user hand-waves RX antenna ~30 cm at ~1 swing/sec, paper Sec. 5.4	136/300 = 45.3% pkt rate -- motion induces delay-Doppler diversity	302	49.3

Plus tx_iq/ — pre-built TX waveforms (cyclic playback inputs).

File format

Each cap_*.npz contains:

rxw (complex64) — raw RX samples (one capture window, length depends on RX script)
meta_json (str) — per-capture metadata as JSON: cfo_hz, pilot_rx, snr_post, stf_idx, etc.
(OFDM step10 captures only) Yeq_data (complex64, shape (num_syms, 48)) — equalised data subcarrier symbols ready for LDPC post-decode without re-running sync/equalisation.
(OFDM step10 captures only) H (complex64) — LTF channel estimate

Filename suffixes: _ok (packet decoded successfully), _bg (background-noise window, no signal), _fail (signal present, decode failed). Each run dir also has a run_summary.json and a rx_config.json (where applicable).

Reproducing the paper

Re-decode the OFDM closed-grid 0 dB run through the LDPC post-decoder:

python3 OTFS/ofdm_ldpc_postdecode.py \
    --rx_npz_glob 'closed_grid_0db_ofdm/cap_*.npz' \
    --ref_seed 42 --ref_len 4096 --chunk_bytes 64 \
    --ldpc_k 256 --ldpc_n 512 --num_ofdm_syms 16 --payload_bytes 64
# Expected: 161 / 489 attempts = 32.9% pkt rate

License

CC-BY-4.0. Please cite the paper if you use this dataset.

@inproceedings{liu2026otfs,
  title  = {Hardware-Validated {OTFS}+{LDPC} on {PlutoSDR}:
            Synthetic Doppler Injection and Monostatic-Loopback Mobility Sensing},
  author = {Liu, Kaikai},
  booktitle = {Proc. IEEE Globecom},
  year   = {2026}
}

Downloads last month: 2

Total file size:

3 GB