This project is a specialized Python-based tool designed for Ferroelectric Random Access Memory (FeRAM) characterization. It automates the processing of raw measurement data (from Keysight B1500A/Keithley 4200), utilizing the PUND algorithm to extract accurate ferroelectric properties while compensating for leakage currents and signal drift.
-
PUND Correction: Automatically identifies
$P, U, N, D$ pulses with adaptive thresholding to remove non-ferroelectric components. - Leakage Compensation: Implements linear compensation to fix "spiraling" hysteresis loops caused by resistive leakage.
- Cycle Isolation: Smartly detects and isolates single-period waveforms from continuous measurement streams.
-
P-E Hysteresis Loops: Overlay multiple cycles (e.g.,
$10^0$ to$10^9$ ) with customizable color maps (Viridis, Plasma, etc.). -
Endurance Trends: Automatically extracts and plots
$2P_r$ (Remnant Polarization) and$2V_c$ (Coercive Voltage) over logarithmic cycle scales. - Data Table: Instant view of key metrics for selected cycles.
- Gemini Assistant: Built-in integration with Google Gemini AI to analyze fatigue, wake-up effects, and potential breakdown mechanisms based on the extracted data.
- Clone the repository:
pip install -r requirements.txt