ATE

Automated test engineering utilities focused on audio and RF experimentation.

Multi-channel NBFM transmitter

This repository includes a multi-channel narrowband FM transmitter that can drive HackRF, Pluto SDR, and PlutoPlus SDR devices. The Python flowgraph in multich_nbfm_tx.py lets you queue playlists per-channel, apply per-channel gains, and stream the composite RF waveform out an osmosdr sink.

High-level workflow

graph TD
    A[Audio playlists per channel] -->|Queued by| B(QueuedAudioSource blocks)
    B -->|Resampled to match target rate| C[Resampler]
    C -->|FM modulated| D[NBFM Modulator]
    D -->|Upconverted & summed| E[Channel Combiner]
    E -->|Sent to SDR| F[osmosdr Sink]
    F -->|RF Transmission| G((Air))

Loading

1. Each logical channel keeps a playlist of WAV and/or MP3 files.
2. QueuedAudioSource reads the playlists, decodes audio, and performs optional sample-rate conversion so every stream matches the transmitter's base audio sample rate.
3. Individual channel audio is frequency shifted to its assigned offset or direct RF frequency, then modulated as narrowband FM.
4. All modulated channels are summed and forwarded to an osmosdr sink which pushes complex baseband samples to the connected SDR at the configured transmit sample rate and gain.

Provided entry points

• multich_nbfm_tx.py – command-line transmitter that streams per-channel file queues (comma-separated file lists), optionally resamples mismatched audio sample rates on the fly, exposes per-channel gain trims, looping control, and either baseband offsets or direct frequency entry (--freqs). Additional channel features include:
  • Optional CTCSS tone generation per channel with either a normalized level (--ctcss-level) or explicit deviation target (--ctcss-deviation).
  • Optional DCS (CDCSS) code injection per channel so you can key receivers that expect digital squelch signaling.
  • Audio-driven gating (--gate-open, --gate-close, --gate-attack-ms, --gate-release-ms) that mutes quiet segments before modulation so idle channels do not inject unnecessary noise into the composite RF signal.
  • Per-channel configuration summaries printed at start-up for quick verification of offsets, gains, tone settings, and estimated bandwidth.
• multich_gui.py – lightweight Tkinter GUI that wraps the transmitter so you can configure devices, FRS/GMRS channel selections, per-channel playlists, and gain levels without memorizing CLI arguments. Core transmitter defaults now live in transmitter_settings.json so you can tweak them once via the GUI's Settings → Manage Transmitter Settings… dialog instead of retyping CLI overrides every launch. Pass python multich_gui.py --help to temporarily override any of the saved defaults via CLI flags.
• ctcss_channel1_squelch.py – helper utility that reuses the main transmitter pipeline to broadcast a continuous CTCSS tone (default 67.0 Hz) on channel 1. This is handy for validating that handheld receivers open squelch for the configured tone without re-entering full playlists.
• audio_activity_report.py – utility that scans WAV/MP3 files and reports their audio duty cycle (percentage of chunks whose RMS amplitude exceeds a configurable threshold). By default it evaluates one-second chunks with a 0.1 RMS threshold and writes the CSV summary to audio_duty_cycle.csv, but you can override any of those settings via CLI flags. Use this to build CSV summaries of long playlists before scheduling them for transmission.

Channel presets

The GUI reads channel_presets.csv (bundled in this repository) to populate its channel picker. Edit that CSV to add, rename, or reorder channels for your site. Each preset must provide a frequency_hz column, and the display_name is shown in the dropdown list.

Transmitter defaults

transmitter_settings.json stores the validated baseline values for TX sample rate, modulation rate, FM deviation, master scale, tone levels, and squelch gate thresholds. Use the GUI's Settings menu to edit these values through a compact dialog. The dialog persists your changes back to the JSON file so subsequent sessions inherit the tuned defaults without exposing the controls on the main window.

Dependencies

Both scripts require GNU Radio with osmosdr support and NumPy available in your Python environment. MP3 playlists are supported via the optional audioread dependency; install it with pip install audioread if you want to mix MP3 tracks alongside WAV files.