Audio-Based Authentication System for Access Control using Boolean FPGA

This project implements a secure, touchless access control system using voice-based authentication on a Boolean FPGA (Spartan-6). It combines Python-generated audio and Verilog-based FPGA design to analyze and compare dominant audio frequencies, allowing access only when the voice patterns match.

---

🚀 Project Overview

• Technology: Spartan-6 FPGA (Boolean Board), Verilog HDL, Python (gTTS, pydub), Vivado Design Suite
• Function: Uses FFT-based frequency analysis to compare a test voice against a stored reference and activates a servo motor if they match (simulates door unlock).
• Applications: Smart home access, secure labs, voice-based authentication systems

---

🎯 System Workflow

🔉 Audio Generation (Python)

• Generates speech using Google Text-to-Speech (gTTS)
• Converts audio to 16-bit PCM and saves as .mem format

🔄 Data Transfer (Python UART)

• Sends chunks via serial to FPGA
• Command protocol:
  • 'R' → Reference
  • 'T' → Test
  • 'D' → Display/Compare

🧠 FPGA Processing (Verilog)

• Stores .mem data into BRAM
• Simulates FFT by finding peak amplitude
• Compares frequencies within a 5% error margin
• Controls LEDs and Servo PWM based on result

🟢 Servo Behavior

• ✅ Match: 90° (Unlocked) → All LEDs ON
• ❌ Mismatch: 0° (Locked) → All LEDs OFF

---

🧠 Key Modules

• uart_rx.v: UART receiver
• uart_bram_led_servo.v: Main FSM, BRAM, FFT-mock logic, PWM generation
• servo_pwm.v: PWM signal for SG90 servo control
• fft_freq_compare.v: (optional advanced) frequency bin comparison

---

💻 Tools Used

• Vivado Design Suite 2024.1 (Synthesis, Simulation, Implementation)
• Python: gTTS, pydub, numpy, pyserial
• Hardware: Boolean FPGA (Spartan-6), SG90 servo, USB-to-UART cable