Project Overview
Led a technical study to rigorously characterize the performance of an on-board 12-bit Analog-to-Digital Converter (ADC). The project focused on quantifying the ADC's noise floor, sensitivity, and Dynamic Range through a structured series of statistical tests to determine its suitability for high-precision robotic sensing.
Technical Responsibilities & Methodologies
Statistical Parameter Modeling: Calculated the Least Significant Bit (LSB) and predicted theoretical Signal-to-Noise Ratio (SNR) based on 12-bit datasheet specifications.
Automated Data Acquisition: Developed custom Python scripts to automate high-frequency data collection, ensuring precise voltage sampling across a series of controlled input increments.
Signal Quality Metrics: Analyzed raw datasets to derive critical performance indicators, including Voltage RMS, Peak-to-Peak noise, and the Effective Number of Bits (ENOB).
Non-Linearity Mapping: Performed rigorous testing to evaluate voltage sensitivity and noise floor characteristics, identifying specific architectural factors that influenced signal fidelity.
Key Achievements
Performance Characterization: Successfully derived an experimental sensitivity that exceeded theoretical expectations, suggesting optimized resolution through internal architecture or averaging techniques.
Correlation Analysis: Identified a moderate linear correlation ($R^2 = 0.566$) between input and output, providing a quantitative baseline for the noise-related factors affecting system accuracy.
Foundation for Improvement: Delivered a comprehensive performance profile that established a clear technical foundation for future ADC filtering and signal conditioning improvements.