STM32 Embedded Filtering & Visualization

This project implements a real-time data acquisition and filtering system using an STM32 microcontroller and a Python-based visualization tool. It demonstrates how to implement complex digital signal processing (DSP) algorithms on a low-memory microcontroller while maintaining high performance.

Project Structure

• STM32 Source/: Contains the firmware for the STM32L031K6Tx microcontroller.
  • Core/Src/main.c: Main application logic.
  • Core/Src/filter.c: Implementation of Butterworth filters (LPF, HPF, BPF, BSF) using optimized math approximations.
• readSTM.py: A Python script to read serial data from the STM32, plot the real-time signal, and display the Frequency Spectrum (FFT).
• signal_generator.ino: Arduino sketch for generating test signals.

Features

• Multi-Type Embedded Filtering: Supports Low Pass (LPF), High Pass (HPF), Band Pass (BPF), and Band Stop (BSF) Butterworth filters.
• Efficient Low-Memory Implementation: Optimized for MCUs with limited resources, using approximation algorithms to avoid heavy standard library dependencies.
• Real-Time Visualization: Live plotting of ADC data using Matplotlib.
• Frequency Analysis: Real-time FFT display to analyze signal frequency components.

Technical Implementation

The system is designed for deterministic real-time performance:

• Sampling Protocol: Uses a hardware timer interrupt to trigger sampling at exactly 1 kHz, ensuring precise signal reconstruction.
• ADC with DMA: The Analog-to-Digital Converter (ADC) operates in DMA Circular Mode. This allows data to be transferred directly to memory without CPU intervention, freeing up the processor for filtering calculations.
• UART Communication: Data transmission is handled via UART with Interrupts, preventing blocking delays during serial communication.
• Robustness: Includes comprehensive Error Handlers within the HAL (Hardware Abstraction Layer) to manage peripheral failures gracefully.

Hardware Specifications

The project is built on the NUCLEO-L031K6 development board, featuring the ultra-low-power STM32L031K6T6 microcontroller.

Microcontroller (STM32L031K6T6)

• Core: ARM® Cortex®-M0+ 32-bit
• Clock Frequency: Up to 32 MHz
• Flash Memory: 32 KB
• RAM: 8 KB
• EEPROM: 1 KB
• Supply Voltage: 1.65 V to 3.6 V
• I/O: 25 GPIOs
• Peripherals: I²C, SPI, UART, USART, and 10-channel ADC

Board Features (NUCLEO-L031K6)

• Form Factor: 32-pin package, Arduino™ Nano V3 compatible.
• Debugger: On-board ST-LINK/V2-1 debugger/programmer with Virtual COM port.
• Power: Flexible options (USB VBUS, external source).
• User Interface: 1 User LED, 1 Reset push-button.
• Clock: 24 MHz crystal oscillator.
• Ecosystem: Arm® Mbed Enabled™ compliant.

Getting Started

Prerequisites

• Hardware: STM32L031K6Tx board (or compatible), USB-UART bridge (if not built-in).
• Software:
  • STM32CubeIDE (for firmware).
  • Python 3.x with pyserial, matplotlib, and numpy.

Running the Python Interface

1. Install dependencies:

   pip install pyserial matplotlib numpy

2. Connect the STM32 board to your computer.
3. Run the script:

   python readSTM.py

   Note: You may need to adjust SERIAL_PORT in readSTM.py to match your system (e.g., COM3 on Windows or /dev/ttyUSB0 on Linux).

Contributors

• @200dollarrbill