Mixture of Gaussian (MoG) Fading Channel Simulator

This project implements a Mixture of Gaussian (MoG) approach to approximate the Probability Density Functions (PDF) of various fading channels (Kappa-Mu, Eta-Mu, Kappa-Mu Shadowed). It validates the theoretical derivations against simulation data and calculates key metrics like Outage Probability.

Project Structure

• main.py: The main entry point for running simulations, generating data, and plotting figures.
• channel_utils.py: The core library containing PDF definitions, MoG fitting logic (EM algorithm, BIC selection), and metric calculations.
• GoM_Paper_Implementation_Slides.pdf: Detailed theoretical background and analysis report.

Requirements

Install the required Python libraries:

pip install -r requirements.txt

Usage

You can run the simulator using the command line interface (CLI).

1. Run the Reproduction Test

Validates if the code can accurately recover the parameters from Table IV of the reference paper.

python main.py --test

2. Plot Specific Figures

To generate specific figures (e.g., Figure 3 for PDF approximation and Figure 5 for Outage Probability):

python main.py --fig 3 5

Available Figures:

• 1: Normalized BIC analysis
• 2: Optimal Components vs Amount of Fading (AF)
• 3: MoG Approximation vs Theoretical PDF
• 5: Outage Probability Analysis

3. Run Full Simulation

To run all simulations, save logs, and plot all figures:

python main.py --all

Output

The simulation results (fitted parameters, KL divergence, MSE, etc.) will be automatically saved to execution_results.txt.

References

This project is a Python implementation based on the mathematical models and parameters described in the following paper:

• Modeling and Analysis of Wireless Channels via the Mixture of Gaussian Distribution
  • Bassant Selim, Omar Alhussein, Sami Muhaidat, George K. Karagiannidis, and Jie Liang
  • IEEE Transactions on Vehicular Technology, Vol. 65, No. 10, pp. 8309-8321, Oct. 2016.

Specifically, this simulation reproduces the analytical results and plots corresponding to Table IV and Figures 1, 2, 3, 5 of the paper to validate the MoG approximation accuracy.

---

Disclaimer: This repository is for educational and interview demonstration purposes only.