ZORONTO Defense AI Repository: Synthetic C4I Simulation Dataset (Cognitive Multi-Agent Decision Optimization in C4I Systems: A Reinforcement Learning and Large Language Model Integration Framework)

This repository provides the full experimental dataset, results, and figures supporting the study "AI-Driven Decision Optimization in C4I Systems: Reinforcement Learning Architecture and Monte Carlo Simulation Analysis"(Journal of Information Science and Engineering, 2025). The repository includes:1. Raw and structured simulation data (Phases 0–4) - 1_synthetic_episode_data_expanded.csv - 2_synthetic_nodes_updated.csv - 3_synthetic_experiment_config_updated.csv - 4_synthetic_links_updated.csv - 5_synthetic_target_trajectory.csv - 6_synthetic_mobility_waypoints.csv - 7_synthetic_jamming_schedule.csv2. Experimental results and summaries (Phases 3–6) - 8_result_episode_metrics_summary.csv - 9_result_phase3.csv - 10_result_phase4.csv - 10_result_phase4_comm_llama3_summary.csv - 11_result_phase5_episode_metrics.csv - 11_result_phase6_episode_metrics.csv - 11_result_phase56_summary.csv3. Figures (Main & Appendix) - Fig_Archi.pdf, Fig2_scenario.png, Fig3_performance_stability_value.png, Fig4_phase4_finegrid_llama3.png, Fig5_phase6_adaptive_llm_nogrid.png, Fig6_convergence_scalability.png, Fig7_pareto_evolution_rigorous.png - FigA1_pipeline6.png, FigA2_ekf_tracking_validation_fixed.png, FigA3_latency_success_mors.png4. Supporting material - baseline_metrics_table_CI.tex - Configuration: environment.yml, config_phase5.yml All simulations were conducted in Python 3.10.13 with PyTorch 2.9.0+CUDA 12.8 (cuDNN 8.9)on an NVIDIA RTX 5090 GPU under WSL2 (Linux 6.6.87.2). Random seeds were fixed andSHA256 checksums are included for all files. This repository ensures full reproducibility of theresults reported in the published manuscript.