AetherSDR Third-Party Software Licenses
========================================

AetherSDR incorporates or links against the following third-party software.
Each component retains its original license. Full license texts are included
in the corresponding third_party/ subdirectories where the source is bundled.


Protocol References
-------------------
AetherSDR implements the SmartSDR protocol to communicate with FlexRadio
transceivers. The protocol behavior was determined by consulting the FlexLib
API source code, which is Copyright FlexRadio Systems and All Rights Reserved.
No FlexLib code is incorporated into AetherSDR — the protocol implementation
is entirely original. SmartSDR is a trademark of FlexRadio Systems.

  FlexLib:   Copyright (c) 2012-2025 FlexRadio Systems
  Reference: https://www.flexradio.com/


1. WDSP — Spectral Noise Reduction (NR2)
-----------------------------------------
The SpectralNR module (src/core/SpectralNR.h, SpectralNR.cpp) is derived from
the WDSP library's emnr.c, implementing the Ephraim-Malah MMSE-LSA noise
reduction algorithm with OSMS noise floor tracking.

  Copyright (C) 2015, 2025 Warren Pratt, NR0V
  License: GPL-2.0-or-later
  Source:  https://github.com/TAPR/OpenHPSDR-wdsp


2. RNNoise — Neural Network Noise Suppression (RN2)
----------------------------------------------------
Bundled at third_party/rnnoise/

  Copyright (c) 2007-2017, 2024 Jean-Marc Valin
  Copyright (c) 2023 Amazon
  Copyright (c) 2017, Mozilla
  Copyright (c) 2005-2017, Xiph.Org Foundation
  Copyright (c) 2003-2004, Mark Borgerding
  License: BSD-3-Clause
  Source:  https://github.com/xiph/rnnoise


3. libspecbleach — Spectral Noise Reduction (NR4)
-------------------------------------------------
Bundled at third_party/libspecbleach/

  Copyright (C) 2022 Luciano Dato <lucianodato@gmail.com>
  License: LGPL-2.1-or-later
  Source:  https://github.com/lucianodato/libspecbleach


4. r8brain-free-src — Sample Rate Converter
--------------------------------------------
Bundled at third_party/r8brain/

  Copyright (c) 2013-2025 Aleksey Vaneev
  License: MIT
  Source:  https://github.com/avaneev/r8brain-free-src


5. RtMidi — MIDI I/O
---------------------
Bundled at third_party/rtmidi/

  Copyright (c) 2003-2023 Gary P. Scavone
  License: MIT (with non-binding request to share modifications)
  Source:  https://github.com/thestk/rtmidi


6. ggmorse — Morse Code Decoder
--------------------------------
Bundled at third_party/ggmorse/

  Copyright (c) 2021 Georgi Gerganov
  License: MIT
  Source:  https://github.com/ggerganov/ggmorse


7. Eclipse Mosquitto — MQTT Client Library
-------------------------------------------
Bundled at third_party/mosquitto/

  Copyright (c) 2010-2021 Roger Light <roger@atchoo.org>
  License: EPL-2.0 OR BSD-3-Clause (dual-licensed)
  Source:  https://github.com/eclipse/mosquitto


8. RADAE — Radio Autoencoder
-----------------------------
Bundled at third_party/radae/

  Copyright (c) 2026 Peter B Marks
  License: BSD-2-Clause
  Source:  https://github.com/drowe67/radae


9. opus-rade — Opus Codec (RADE variant)
-----------------------------------------
Bundled at third_party/opus-rade/

  Copyright 2001-2023 Xiph.Org, Skype Limited, Octasic,
                      Jean-Marc Valin, Timothy B. Terriberry,
                      CSIRO, Gregory Maxwell, Mark Borgerding,
                      Erik de Castro Lopo, Mozilla, Amazon
  License: BSD-3-Clause
  Source:  https://github.com/nicoboss/opus-rade


10. DeepFilterNet — Deep Noise Suppression
-------------------------------------------
Bundled at third_party/deepfilter/

  Copyright (c) DeepFilterNet Contributors
  License: Apache-2.0 OR MIT (dual-licensed)
  Source:  https://github.com/Rikorose/DeepFilterNet


11. ONNX Runtime — Neural Network Inference Engine (optional)
--------------------------------------------------------------
Pre-built binaries at third_party/onnxruntime/ (Windows only; optional feature).
Loaded at runtime for the S-History CNN signal classifier (S_History_v2-AI).
Only present when setup-onnxruntime.ps1 has been run.

  Copyright (c) Microsoft Corporation. All rights reserved.
  License: MIT
  Source:  https://github.com/microsoft/onnxruntime

  Note: the tools/train_classifier.py training script requires PyTorch
  (BSD-3-Clause, https://pytorch.org/) as a separate installation.
  PyTorch is a build tool only — it is not linked into or distributed
  with AetherSDR.  The trained .onnx model file produced by the script
  is entirely the user's own work.


12. FFTW3 — Fast Fourier Transform Library
--------------------------------------------
Pre-built binaries at third_party/fftw3/ (Windows only; Linux uses system pkg)

  Copyright (c) 2003, 2007-2014 Matteo Frigo
  Copyright (c) 2003, 2007-2014 Massachusetts Institute of Technology
  License: GPL-2.0-or-later
  Source:  https://www.fftw.org/


12. TCI Protocol Specification — Transceiver Control Interface
---------------------------------------------------------------
The TCI server (src/core/TciProtocol.cpp, TciServer.cpp) implements the
TCI v2.0 protocol specification published by Expert Electronics. The binary
audio frame header follows the spec's Stream struct definition. No code was
copied from ExpertSDR3; this is an independent implementation of the published
protocol.

  Copyright (c) 2021 Expert Electronics (EE)
  License: MIT
  Spec:    https://github.com/ExpertSDR3/TCI


13. zlib — Compression Library
-------------------------------
Bundled at third_party/zlib/ (v1.3.1) for raw-deflate decompression of
.ssdr_cfg profile import/export ZIP packages.  Vendored from the upstream
GitHub release (sha256
9a93b2b7dfdac77ceba5a558a580e74667dd6fede4585b91eefb60f03b72df23) so the
version is identical on Linux, macOS, and Windows.  Previously, Linux/macOS
linked the system zlib and Windows pulled it via vcpkg; bundling unifies the
three paths per the constitution's Technology Constraint preferring bundled
libraries.  See #2651.

  Copyright (C) 1995-2024 Jean-loup Gailly and Mark Adler
  License: zlib license (permissive, similar to MIT)
  Source:  https://github.com/madler/zlib
