AnalogCIM-CSNR-Sim

This repository contains Python scripts for simulating and analyzing Signal-to-Noise Ratio (SNR) in Analog Compute-In-Memory (CIM) architectures. The simulation focuses on quantization effects and noise in CIM operations, providing insights into the performance of these systems under various conditions.

Features

• Simulation of ACIM calculation process with quantization and noise
• Analysis of SQNR and CSNR under various conditions
• Visualization of results for different ADC bit resolutions and noise levels
• Replication of results from published research papers

Requirements

• Python 3.x
• NumPy
• Matplotlib

Usage

1. Clone the repository:

   git clone https://github.com/Keio-CSG/AnalogCIM-CSNR-Sim
   

2. Navigate to the project directory:

   cd AnalogCIM-CSNR-Sim
   

3. Run 'CIM_CSNR_simulation.ipynb'