Simulation code and data for RSMA/SDMA/NOMA downlink benchmarking

Description

MATLAB simulation code, figures, and result tables supporting the manuscript “Reproducible Benchmarking of RSMA, SDMA, and NOMA Under Unified Multi-User Downlink Assumptions,” submitted to Physical Communication. The repository provides a reproducible benchmarking framework for comparing Rate-Splitting Multiple Access (RSMA), Space-Division Multiple Access (SDMA), and Non-Orthogonal Multiple Access (NOMA) under a unified set of channel, beamforming, and evaluation assumptions. The simulations consider a two-user downlink MISO system with M = 4 base-station antennas, K = 2 single-antenna users, an SNR sweep from 0–30 dB, fixed random seeds, and paired Monte Carlo channel realisations. Full SNR sweeps use 1000–2000 Monte Carlo iterations per SNR point, while targeted per-realisation statistical comparisons use up to 5000 iterations. The dataset includes MATLAB code for scheme-specific precoding and rate calculation, shared configuration and channel-generation modules, validation scripts, MATLAB Online-compatible run scripts, final manuscript figures, and CSV tables underlying the reported results. The included README.md describes the folder structure, simulation parameters, reproducibility instructions, and the mapping between manuscript figures/tables and their corresponding source scripts and data files.

Steps to reproduce

Reproducing the Results 1. Add the `code/` folder, including all subfolders, to your MATLAB path. 2. Shared simulation parameters are defined once in: `code/config/config.m` This file contains the antenna/user configuration, SNR range, power splits, regularisation constants, and energy-efficiency model constants. 3. Run the relevant script for each scheme or scenario. For example: `code/rsma/run_rsma_power_sweep.m` If using MATLAB Online, use the corresponding scripts in: `code/matlab_online_variants/` See that folder’s README for the flat-file upload workflow. 4. Each script generates channel realisations using: `code/config/generateChannel.m` The scripts output rate, fairness, and energy-efficiency statistics corresponding to the CSV files in: `results/tables/` 5. Figures in: `results/figures/` are produced from the corresponding result tables. 6. The top-level `README.md` provides a full mapping between each manuscript figure/table and its source script and data file.

Institutions

• University of Sussex

  England, Brighton

Categories

Licence