AFCI offline kit - Part 1: STM32F4 firmware & embedded libraries

Offline development kit for an arc-fault detection / electrical-fire prevention system (current sensor -> STM32F407 ADC+DMA -> FFT/STFT features -> quantized DNN -> trip). Target offline PC: Windows 10/11 x64, CPU-only Python training.

Contents: STM32F4 firmware & embedded libraries

Folder What Why you need it License
01_STM32CubeF4/ STM32CubeF4 full MCU package (GitHub STMicroelectronics/STM32CubeF4, all submodules): HAL + LL drivers, CMSIS Core/Device, BSPs, middleware (FreeRTOS, USB, FatFs, LwIP), and hundreds of example projects for F4 boards (incl. STM32F4-Discovery / STM32F407). Also contains the HAL user manual (.chm) under Drivers/STM32F4xx_HAL_Driver/. Base firmware for the STM32F407: ADC + DMA + TIM examples (Projects/STM32F4-Discovery/Examples/ADC, .../DMA, .../TIM). Point STM32CubeMX/CubeIDE at this zip via Help > Manage embedded software packages > From Local (note: CubeMX may still prefer the ST-packaged en.stm32cubef4_v1-xx-x.zip, see the checklist in Part 7). BSD-3 / Apache-2.0 / ST SLA0044 (per component)
02_CMSIS/ CMSIS_6 (Core), CMSIS-DSP (FFT, FIR/IIR, matrix, statistics, windowing, optimized for Cortex-M4 DSP/FPU), CMSIS-NN (int8 NN kernels), CMSIS_5 (legacy; same generation as the CMSIS copy inside CubeF4). ADC window -> filter -> FFT -> features on the F407; fast int8 inference. Apache-2.0
03_TinyML_runtime/ TensorFlow Lite for Microcontrollers source (tflite-micro). Alternative to ST Edge AI / X-CUBE-AI runtime: run a quantized .tflite model on the F407 with CMSIS-NN kernels. Apache-2.0
04_Docs_HTML/ Built HTML documentation for CMSIS 6 (v6.4.0), CMSIS-DSP (v1.18.0), CMSIS-NN (v8.0.0). Open index.html. API reference offline. Apache-2.0
05_RTOS/ FreeRTOS-Kernel (latest). CubeF4 also ships its own FreeRTOS copy under Middlewares/Third_Party. Later stage (tasks: acquisition / DSP / inference / safety / logging). MIT

All archives are standard .zip (open with Windows Explorer or 7-Zip). Git history was removed to save space.

Rebuilding split files

Files over 100 MB were split into *.part001, *.part002, ... pieces. After downloading the whole repo:

  • Windows: powershell -ExecutionPolicy Bypass -File .\join_parts.ps1
  • Linux/macOS: bash join_parts.sh

Both scripts rebuild the files and check them against SHA256SUMS.txt.

Files (10 files, 285.6 MB total)

File Size Split into
01_STM32CubeF4/STM32CubeF4_full_with_submodules_2026-10-05.zip 181.6 MB 2
02_CMSIS/CMSIS-DSP_2026-10-28.zip 39.1 MB -
02_CMSIS/CMSIS-NN_2026-11-02.zip 7.3 MB -
02_CMSIS/CMSIS_5_2024-09-03.zip 21.9 MB -
02_CMSIS/CMSIS_6_2026-11-02.zip 9.1 MB -
03_TinyML_runtime/tflite-micro_2026-10-30.zip 11.8 MB -
04_Docs_HTML/CMSIS-DSP_html_docs.zip 3.2 MB -
04_Docs_HTML/CMSIS-NN_html_docs.zip 1.3 MB -
04_Docs_HTML/CMSIS_6_html_docs.zip 6.6 MB -
05_RTOS/FreeRTOS-Kernel_2026-10-29.zip 3.8 MB -

Each file keeps the license of its original publisher (see the notes above).

Master index

This is one of 19 repos (AFCI_part1 โ€ฆ AFCI_part19). Overview of all parts: AFCI_part6 / 00_START_HERE_MASTER_INDEX.md ยท Installation: AFCI_part6 / 00_INSTALLATION_GUIDE.md

Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. ๐Ÿ™‹ Ask for provider support