Nexys 4 DDR PDM-MIC ETHERNET FRAME GENERATOR

Overview

This project involves the Nexys4DDR, which processes data from a built-in microphone using Pulse Density Modulation (PDM). It includes digital filtering for decimation and resolution enhancement, as well as audio data reverberation. Additionally, it performs Fast Fourier Transform (FFT) computations over the Ethernet interface using the 100-base-T standard. The packets are captured on the host machine using RAW_SOCKETS and processed using various signal processing algorithms.

Peripherals Used

• PDM microphone
• 10/100 Ethernet PHY

Description

The Nexys4DDR acquires PDM data from the built-in microphone, applies digital filtering for decimation and resolution (16-bit, 48KSPS), and subsequently adds audio reverberation before outputting it through the built-in Audio Out interface. Following this, the machine computes the Fast Fourier Transform (FFT) of the time-domain signal and transmits it over the Ethernet. On the host side, we capture the packets using RAW_SOCKETS and display the real-time Fourier spectrum or play back the recorded audio signal.

Prerequisites

Hardware

• Nexys 4 DDR FPGA board
• Micro-USB cable
• Ethernet cable

Software

• Vivado Design Suite 2021.1+ (Procedure may vary with other versions)
• Python (latest version)
• Wireshark

Setup Instructions

1. Clone the Repository

   git clone https://github.com/the-pinbo/ddfpga-project.git

2. Generate the Project

   • Open Vivado.
   • Click Run Tcl Script in the tools window.
   • Navigate to ddfpga-project/project and select efsm.tcl.
   • Upgrade IPs if necessary via Reports > Report IP Status.
   • Note: Change MAC address before generating the bitstream.
     • Linux: Use ifconfig -a in terminal.
     • Windows: Use ipconfig /all in Command Prompt.

3. Build the Project

   • Use Generate Bitstream in Vivado.
   • Run Synthesis and Implementation steps.

4. Program the Board

   • Connect the FPGA board.
   • Use Auto-Connect in hardware manager.
   • Load the program with program device.

5. Capture Packets

   • Connect Ethernet from FPGA to computer.
   • Use Wireshark for packet analysis.
   • Save packets as .pcap for processing.

6. Setting Up a Virtual Environment

   • Linux

     python3 -m venv venv
     source venv/bin/activate
     pip install -r requirements.txt

   • Windows

     python -m venv venv
     .\venv\Scripts\activate
     pip install -r requirements.txt

7. Packet Defragmentation and Audio Generation

   This process involves defragmenting packets and performing signal processing to generate an audio file from the captured data.

   • Generating Audio Using Wireshark Analysis Run python\wireshark_analysis.ipynb to analyze the data and generate an audio file in .wav format.
   • Generating Audio Directly (Linux Only) Note: This method requires sudo access.
     1. Open python\receive_wav.py and configure the Ethernet interface and destination MAC address.
     2. Run the receive_wav.py file.

8. Play Back Audio

   • Enjoy the processed voice through the microphone.

For detailed implementation, see Implementation Details.