SynNetQoS: A Transparent Simulation-Based Synthetic 4G/5G Dataset Generator for QoS and QoE Modeling

Description SynNetQoS is a transparent simulation-based synthetic data-generation framework for controlled 4G/5G Quality of Service (QoS) and Quality of Experience (QoE) modeling experiments. The reference release contains 50,000 time-correlated records from 5,000 synthetic user sessions and 67 generated columns covering 4G, 5G non-standalone, and 5G standalone scenarios. The release includes the synthetic dataset, generator code, workflow scripts, result summaries, figures, and documentation. The generator combines a selected 3GPP TR 38.901-inspired UMi-Street Canyon large-scale propagation mode with explicit mobility, congestion, traffic-demand, QoS, QoE, and dropped-connection rules. Reproducibility is supported through fixed generation configuration, deterministic workflow scripts, saved metadata, propagation audit fields, result summaries, and a dataset SHA-256 hash. SynNetQoS is intended as an auditable synthetic research artifact for controlled QoS/QoE modeling, selected external-alignment studies, simulator-reference diagnostics, and leakage-aware machine-learning benchmark design. It is not intended to replace restricted operator traces, field measurement campaigns, packet-level simulators, ns-3, or 5G-LENA, and it should not be interpreted as a field-calibrated operational mobile-network dataset. The overall project directory structure is presented as follows: synnetqos/├─ CITATION.cff├─ LICENSE├─ LICENSE-DATA├─ NOTICE├─ README.md├─ pyproject.toml├─ requirements.txt├─ config/│ ├─ generator_reference.yaml│ └─ paths.example.yaml├─ data/│ ├─ README.md│ ├─ external/│ └─ synthetic/│ └─ synnetqos-dataset.csv├─ figures/│ ├─ README.md│ ├─ external_alignment/│ │ ├─ external_download_ecdf_syn_vs_campus_controlled.pdf│ │ ├─ external_download_ecdf_vienna_phone_matched.pdf│ │ ├─ external_jitter_ecdf_syn_vs_campus_controlled.pdf│ │ └─ external_rsrp_ecdf_vienna_phone_matched.pdf│ ├─ ml_benchmark/│ │ └─ ml_task_result_bars.pdf│ ├─ simulator_comparison/│ │ └─ simulator_throughput_comparison.pdf│ └─ supplementary/│ ├─ correlation_heatmap.pdf│ ├─ dataset_coverage_heatmap.pdf│ ├─ external_rsrp_to_throughput_trend.pdf│ ├─ internal_check_mobility_vs_handovers.pdf│ ├─ internal_check_obstruction_vs_signal.pdf│ ├─ latency_vonr_boxplot.pdf│ ├─ ml_future_drop_precision_recall_curves.pdf│ ├─ ml_streaming_qoe_precision_recall_curves.pdf│ ├─ monte_carlo_stability.pdf│ └─ simulator_delay_jitter_comparison.pdf├─ results/│ ├─ README.md│ ├─ external_alignment/│ │ ├─ controlled_5g_subset_audit.csv│ │ ├─ external_alignment_campus_controlled_5g_summary.csv│ │ ├─ external_alignment_diagnostic_summary.csv│ │ ├─ external_alignment_interpretation_flags.csv│ │ ├─ external_alignment_summary.csv│ │ ├─ external_alignment_vienna_phone_summary.csv│ │ ├─ external_feature_mapping.csv│ │ └─ supplementary/│ │ ├─ external_alignment_vienna_positive_downlink_summary.csv│ │ ├─ external_rsrp_to_throughput_trend.csv│ │ ├─ external_spearman_rsrp_throughput.csv│ │ └─ vienna_scanner_rsrp_summary.csv│ ├─ generator/│ │ ├─ dataset_integrity_summary.csv│ │ ├─ dataset_schema.csv│ │ ├─ drop_event_summary.csv│ │ ├─ generator_config.json│ │ ├─ monte_carlo_run_metrics.csv│ │ ├─ monte_carlo_stability_summary.csv│ │ ├─ numerical_range_summary.csv│ │ ├─ outlier_report.csv│ │ ├─ propagation_model_audit.csv│ │ └─ vonr_latency_consistency.csv│ ├─ ml_benchmark/│ │ ├─ ml_benchmark_run_metrics.csv│ │ ├─ ml_benchmark_summary.csv│ │ ├─ ml_confusion_summary.csv│ │ ├─ ml_feature_importance.csv│ │ ├─ ml_feature_sets.csv│ │ ├─ ml_leakage_audit.csv│ │ ├─ ml_reproducibility_metadata.json│ │ ├─ ml_split_summary.csv│ │ ├─ ml_target_prevalence.csv│ │ └─ ml_task_definitions.csv│ └─ simulator_comparison/│ ├─ simulator_comparison_interpretation_flags.csv│ ├─ simulator_comparison_verdict.csv│ ├─ simulator_feature_mapping.csv│ ├─ simulator_kpi_comparison.csv│ ├─ simulator_kpi_summary.csv│ ├─ simulator_kpi_trend_summary.csv│ ├─ simulator_packet_loss_summary.csv│ ├─ simulator_trace_manifest.json│ └─ simulator_trace_quality_flags.csv├─ scripts/│ ├─ 01_generate_dataset.py│ ├─ 02_internal_consistency_checks.py│ ├─ 03_monte_carlo_stability.py│ ├─ 04_external_alignment.py│ ├─ 05_ml_benchmark.py│ └─ 06_simulator_comparison.py├─ simulators/│ └─ ns3_lena/│ └─ run_cttc_nr_demo_grid.sh├─ src/│ └─ synnetqos/│ ├─ __init__.py│ ├─ audits.py│ ├─ config.py│ ├─ generator.py│ ├─ io.py│ ├─ ml.py│ ├─ plotting.py│ ├─ profiles.py│ ├─ simulator_comparison.py│ └─ validation.py└─ tests/ ├─ test_generator_kpi_scaling.py ├─ test_integrity.py └─ test_simulator_comparison.py