📡 Arduino Fast Fourier Transform (FFT) Analyzer

Real-time signal processing and frequency analysis using the Adafruit ZeroFFT library.

This project provides a robust implementation of the Fast Fourier Transform (FFT) algorithm on Arduino-compatible microcontrollers. It is designed to analyze signals in real-time, converting time-domain data into frequency-domain components to identify dominant frequencies. Whether you are analyzing audio, vibration, or synthetic signals, this tool offers a straightforward way to visualize and compute spectral data.

---

🚀 Key Features

• Real-time Signal Processing: Capable of reading analog signals directly from an input pin (A0) and processing them on the fly.
• High-Speed FFT: Utilizes the optimized Adafruit_ZeroFFT library for efficient computation on 32-bit microcontrollers (e.g., SAMD21/SAMD51).
• Dual Operation Modes:
  • Live Mode: Analyzes real-world signals from sensors or audio inputs.
  • Test Mode: Uses hard-coded signal data (synthetic sine waves) for debugging and verification without external hardware.
• Dominant Frequency Detection: Automatically calculates and outputs the frequency with the highest magnitude.
• Visualization Ready: Formatted output compatible with the Arduino Serial Plotter for real-time graphical visualization of the spectrum.

---

🛠 Tech Stack

Component	Technology / Protocol	Description
Language	C++	Core implementation language for Arduino sketch.
Framework	Arduino SDK	Development platform and HAL.
Library	Adafruit_ZeroFFT	Optimized FFT library for ARM Cortex-M0/M4.
Hardware	Arduino (SAMD)	Recommended hardware (e.g., Arduino Zero, Feather M0).
Input	Analog ADC	signal acquisition via pin A0.
Output	UART / Serial	Data streaming at 9600 baud.

---

🏗 Architecture & Design

The system operates on a cyclic processing model designed for continuous analysis:

1. Data Acquisition Phase:
   • The system collects SAMPLE_SIZE (default: 256) data points.
   • In Live Mode, it samples the voltage at pin A0.
   • In Test Mode, it loads pre-computed sine wave values from signal.h.
2. Processing Phase (FFT):
   • The time-domain signal is passed to the ZeroFFT algorithm.
   • The algorithm converts the signal into frequency bins.
3. Analysis Phase:
   • The system iterates through the frequency bins (up to SAMPLING_RATE / 2).
   • It calculates the magnitude for each bin and tracks the maximum value to identify the Dominant Frequency.
4. Output Phase:
   • Results (Frequency vs. Amplitude) are sent over Serial for logging or plotting.

Note: The code uses a "blocking" read mechanism for simplicity, ensuring signal continuity during the sampling window.

---

🔌 Prerequisites & Installation

Hardware Requirements

• Arduino Zero, Feather M0, or any SAMD21/SAMD51 based board. (Note: Adafruit_ZeroFFT is optimized for these architectures).
• Signal source (Function generator, Microphone amplifier, or simple wire for noise testing).

Software Setup

1. Install Arduino IDE: Download from arduino.cc.
2. Install Library:
   • Open Arduino IDE.
   • Go to Sketch -> Include Library -> Manage Libraries...
   • Search for Adafruit ZeroFFT and install the latest version.

Installation Steps

1. Clone this repository:

   git clone https://github.com/your-username/FFT-Arduino-Project.git

2. Open the FFT.ino file in the Arduino IDE.
3. Select your board and port in Tools -> Board / Port.

---

💻 Usage

1. Configuration

Open FFT.ino and modify the configuration macros at the top of the file to suit your needs:

// Set to 1 to use internal test data, 0 for real sensor data
#define USE_HARD_CODED_DATA 0 

// Set the sample rate (must match your actual sampling frequency)
#define SAMPLING_RATE 80 

// FFT Size (Power of 2, e.g., 256, 512)
#define SAMPLE_SIZE 256 

2. Running the Analyzer

1. Connect your analog signal source to Pin A0 (if using USE_HARD_CODED_DATA 0).
2. Upload the sketch to your board.
3. Open the Serial Plotter (Ctrl+Shift+L) to see the frequency spectrum visually.
4. Alternatively, open the Serial Monitor (Ctrl+Shift+M) at 9600 baud to see textual data and the calculated dominant frequency.

Example Output

30 Hz: 12.50
31 Hz: 45.10
...
Dominant frequency: 31.00 Hz

---

📊 MATLAB Verification

This project includes a MATLAB script to verify the accuracy of the Arduino FFT implementation. You can use it to compare the on-board processing results with a high-precision reference.

Files

• Matlab_code/Arduino_FFT_Spectral.m: The main script for offline FFT analysis.
• Matlab_code/Input_Signal.mat: Pre-recorded test signal data.
• Matlab_code/Input.txt: Text version of the signal data.

How to Run

1. Open MATLAB.
2. Navigate to the Matlab_code/ directory.
3. Load the signal data:

   load('Input_Signal.mat'); % Loads 'Input_Signal' variable into workspace

4. Run the validation script:

   Arduino_FFT_Spectral

5. A plot will appear showing the Single-Sided Amplitude Spectrum. Compare the peak frequency and amplitude distribution with your Arduino Serial Plotter results to validate the implementation.

Note: The MATLAB script assumes a sampling frequency (Fs) of 8000 Hz (matching the data generation). Ensure your Arduino implementation is configured correctly to match or account for this when comparing.

---

🔮 Future Improvements

Here is the roadmap for future development:

• Dynamic Sampling Rate: Implement interrupt-based sampling for more precise timing.
• OLED/LCD Integration: Display the dominant frequency directly on a screen.
• Noise Filtering: Add software filters (e.g., Moving Average) before FFT processing.
• Windowing Functions: Implement Hanning or Hamming windows to reduce spectral leakage.
• Optimized Storage: Use DMA (Direct Memory Access) for signal acquisition to free up CPU time.

---

Created by [Your Name]