Skip to content

RFlect

API

RFingAdam/RFlect

API Reference¶

Pure-Python entry points you can import from plot_antenna.

`AntennaAnalyzer` — `plot_antenna.ai_analysis`¶

GUI-independent analysis engine. Same code path used by the AI chat assistant and the MCP analysis tools.

from plot_antenna.ai_analysis import AntennaAnalyzer

data = {
    "phi":         phi_angles,      # 1-D array
    "theta":       theta_angles,    # 1-D array
    "total_gain":  gain_array,      # 2-D or flattened
    "h_gain":      h_pol_array,
    "v_gain":      v_pol_array,
}

analyzer = AntennaAnalyzer(
    measurement_data=data,
    scan_type="passive",            # or "active"
    frequencies=[2400.0, 2450.0],
)

stats   = analyzer.get_gain_statistics(frequency=2450.0)
pattern = analyzer.analyze_pattern(frequency=2450.0)
polz    = analyzer.compare_polarizations(frequency=2450.0)

File parsers — `plot_antenna.file_utils`¶

Function	Returns
`read_passive_file(path)`	`(parsed_data, start_phi, stop_phi, inc_phi, start_theta, stop_theta, inc_theta)`
`read_active_file(path)`	Dict with `Frequency`, axis ranges, `H_Power_dBm`, `V_Power_dBm`
`parse_2port_data(path)`	DataFrame with `! Stimulus(Hz)`, `S11(dB)`, `S21(dB)`, `S21(s)`
`parse_touchstone_to_dataframe(path)`	Same DataFrame shape from `.s2p`
`batch_process_passive_scans(...)`	Process every HPOL/VPOL pair in a folder
`batch_process_active_scans(...)`	Process every TRP file in a folder
`convert_HpolVpol_files(...)`	Export to CST `.ffs`
`validate_hpol_vpol_files(...)`	`(bool, error_msg)` tuple

Calculations — `plot_antenna.calculations`¶

Function	Returns
`calculate_passive_variables(...)`	Total/H/V gain arrays + theta, phi grids
`calculate_active_variables(...)`	TRP, H/V TRP, total power 2D, min/nom stats
`extract_passive_frequencies(path)`	List of MHz available in a passive file

UWB — `plot_antenna.uwb_analysis`¶

Function	Returns
`calculate_sff(freq, s21, pulse_type)`	SFF + quality label + peak delay
`calculate_sff_vs_angle(angle_data, pulse_type)`	Per-angle SFF + mean
`compute_group_delay_from_s21(freq, s21)`	Group delay + variation + distance error
`extract_transfer_function(freq, s21, distance_m)`	`H(f)` mag/phase + flatness
`compute_impulse_response(freq, H_complex)`	Pulse width + ringing
`analyze_return_loss(freq, s11_dB, threshold_dB)`	Bandwidth + band edges + FBW %
`parse_touchstone(path)`	Dict with `freq_hz`, `s11`, `s21`

Calibration drift — `plot_antenna.cal_drift`¶

Function	Purpose
`record_run(cal_result, ...)`	Record one calibration run with metadata
`import_historical_dir(path)`	Walk + record every `TRP Cal *.txt` in dir
`list_runs(antenna, band)`	DataFrame of recorded runs
`compute_drift(baseline_id, current_id)`	Drift result object
`result_to_dict(result, max_delta_rows)`	JSON-safe dict
`export_markdown(result, path)`	Markdown drift report
`export_pdf(result, path)`	PDF drift report
`render_delta_plot(result, out_path)`	PNG drift plot
`history_dir()` / `set_history_dir(path)`	Where history lives
`set_setup_group(run_id, group)`	Tag a run's methodology epoch
`update_notes(run_id, text)`	Operator notes

LLM provider — `plot_antenna.llm_provider`¶

from plot_antenna.llm_provider import get_provider, UnifiedMessage, ToolDefinition

provider = get_provider("openai")  # or "anthropic", "ollama"
response = provider.chat(
    messages=[UnifiedMessage(role="user", content="...")],
    tools=[ToolDefinition(name="get_gain_statistics", ...)],
)

See plot_antenna/llm_provider.py for BaseLLMProvider, LLMResponse, ToolCall.

API key management — `plot_antenna.api_keys`¶

from plot_antenna.api_keys import store_key, get_key, delete_key

store_key("openai", "sk-...")
key = get_key("openai")        # returns plaintext from keyring or Fernet store
delete_key("openai")

Keys are AES-128 (Fernet) encrypted at rest with PBKDF2-HMAC-SHA256 (600 K iterations) and a machine-ID-derived salt. Storage priority: OS keyring → Fernet file → environment variable → .env.