A Framework for Emerging Nanoscale Memory Characterization using the ArC TWO platform

This repository contains the full implementation, documentation, and results of a semester project focused on developing a flexible framework for memristive device and crossbar array characterization using the ArC TWO platform.

🧠 Project Overview

As AI and neuromorphic computing demand new memory technologies, characterizing memristive devices efficiently and reliably becomes essential. This project implements a modular Python-based measurement framework using the ArC TWO board — a 64-channel FPGA platform tailored for analog and digital memory testing.

The system enables:

• IV sweeps
• Arbitrary pulse-based protocols (e.g., LTP/LTD, Retention, Endurance)
• Crossbar array readout with sneak-path mitigation
• Data logging in HDF5
• Jupyter notebook-based experimentation

📁 Repository Structure

├── src/
│   ├── analysis/             # Plotting & data visualization
│   ├── instruments/          # Pulse sequence generation & device interface
│   ├── measurement/          # IV, pulsing, crossbar logic
│   └── utils/                # Data I/O, hardware utilities
├── examples/
│   └── main.ipynb            # Entry-point notebook for experiments
├── crossbar_config/          # TOML files for pin mapping
├── docs/                     # Documentation and guides
│   ├── WIP_ArC_TWO_Guide_V4.pdf  # ETH Zurich user setup guide
│   ├── Semester_Project_Report.pdf # Report
│   └── Semester_project_presentation_Jeff_Ren.pdf # Presentation slides
└── README.md                 # This file

🚀 Getting Started

Requirements

• Python 3.8+
• pyarc2 (Python bindings for ArC TWO)
• numpy, matplotlib, h5py, etc.
• Jupyter Lab or Notebook
• check 📄 WIP ArC TWO Guide (PDF)

Running an Experiment

Open the notebook:

cd examples
jupyter notebook main.ipynb

The notebook demonstrates:

• How to connect to the ArC TWO board
• IV sweep configuration and execution
• Arbitrary pulsing (LTP/LTD, retention, endurance)
• Crossbar array interaction with mock data

🧑‍💻 ETH Zurich Lab Setup

If you're working from ETH lab computers (e.g. steghorn), setup instructions, conda environment configuration, firmware flashing, and remote GUI usage are documented in the following guide:

📄 WIP ArC TWO Guide (PDF)

📝 Report

The final semester report and presentation slides are included in:

• /docs/Semester_Project_Report.pdf
• /docs/Semester_project_presentation_Jeff_Ren.pdf

📚 References

Key technologies and concepts:

• Memristive devices (VCM, FTJ)
• In-memory computing and neuromorphic hardware
• ArC TWO platform by ArC Instruments
• Pulse-based characterization protocols

🤝 Acknowledgements

This project was conducted as part of the Spring 2025 Semester Project at ETH Zurich under the supervision of:

• Till Zellweger
• Dr. Nikhil Garg
• PD. Dr. Alexandros Emboras
• Prof. Dr. Laura Bégon-Lours
• Prof. Dr. Mathieu Luisier

---