Gespeichert in:
| Hauptverfasser: | , |
|---|---|
| Format: | Recurso digital |
| Sprache: | |
| Veröffentlicht: |
Zenodo
2025
|
| Online-Zugang: | https://doi.org/10.5281/zenodo.16265621 |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866902236520513536 |
|---|---|
| author | Yao, Xiao PEREZ YUSTE, ANTONIO |
| author_facet | Yao, Xiao PEREZ YUSTE, ANTONIO |
| contents | <p>A machine learning (ML) based energy optimization framework for 5G New Radio (5G NR) utilizing a Classification and Regression Tree (CART) algorithm was proposed by the authors of this dataset. We designed a statistically grounded UE traffic model based on a custom RRC State Diagram to simulate variant UE demand. As a result, a Traffic Model dataset was produced consisting of: </p> <p>- One PNG file, indicating the 5G RAN testbed we designed for our research and dataset generation. In this layout, an inter-band NR-NR Dual Connectivity (DC) RAN architecture are proposed. The n28, n78 and n258 bands in the layout are frequency bands defined in the 5G NR (New Radio) standard, which is respectively located at 700 MHz, 3.5GHz and 26GHz.<br>- One compressed archive (`UEdemand.zip`) containing 21,504. Each CSV file inside the ZIP archive is named with the UE distribution scenario it simulates. Each CSV file contains the UE traffic data we simulated under this distribution scenario, traffic data including: DRB demand in each cell, UE thourghput in each cell, throughput for different type of UE in each cell. <br>- Two CSV documents (Extracted99PercentileofDrbDemand.CSV and MatechedLayerandState.CSV). For Extracted99PercentileofDrbDemand.CSV, it contains the UE distribution and the 99-percentile value of the DRB demand in each cell, with the data format of:[UserNumCase1, UserNumCase2, UserNumCase3, UserNumCase4, 99-percentile of DRB demand in n28, 99-percentile of DRB demand in n78-1, 99-percentile of DRB demand in n78-2, 99-percentile of DRB demand in n78-3, 99-percentile of DRB demand in n258]. For MatechedLayerandState.CSV, it contains the UE distribution and the matched sleep state in each cell, with the data format of: [UserNumCase1, UserNumCase2, UserNumCase3, UserNumCase4, LxSy in n28, LxSy in n78-1, LxSy in n78-2, LxSy in n78-3, LxSy in n258]. In these dataset, the UserNumCase1, 2, 3 and 4 are the UE population respectively in n258 cell, n78-2 cell, n78-2 cell, n78-3 cell and n28 cell; the LxSy represents different level of Sleep States we assigned to each cell, L represents the layer, S represents the State, LxSy represents the sleep state of Layer x and Sate y. </p> <p>The dataset is designed for machine learning and telecommunications research.</p> <p>More details of how the dataset is simulated and aplicable scenarios can be found in paper: "A ML-Based Resource Allocation Scheme for Energy Optimi-zation in 5G NR", MDPI sensor, 2025.</p> |
| format | Recurso digital |
| id | zenodo_https___doi_org_10_5281_zenodo_16265621 |
| institution | Zenodo |
| language | |
| publishDate | 2025 |
| publisher | Zenodo |
| record_format | zenodo |
| spellingShingle | RRC State Diagram 5G-NR UE Traffic Model Dataset (5G-RRC-SD-UE) Yao, Xiao PEREZ YUSTE, ANTONIO <p>A machine learning (ML) based energy optimization framework for 5G New Radio (5G NR) utilizing a Classification and Regression Tree (CART) algorithm was proposed by the authors of this dataset. We designed a statistically grounded UE traffic model based on a custom RRC State Diagram to simulate variant UE demand. As a result, a Traffic Model dataset was produced consisting of: </p> <p>- One PNG file, indicating the 5G RAN testbed we designed for our research and dataset generation. In this layout, an inter-band NR-NR Dual Connectivity (DC) RAN architecture are proposed. The n28, n78 and n258 bands in the layout are frequency bands defined in the 5G NR (New Radio) standard, which is respectively located at 700 MHz, 3.5GHz and 26GHz.<br>- One compressed archive (`UEdemand.zip`) containing 21,504. Each CSV file inside the ZIP archive is named with the UE distribution scenario it simulates. Each CSV file contains the UE traffic data we simulated under this distribution scenario, traffic data including: DRB demand in each cell, UE thourghput in each cell, throughput for different type of UE in each cell. <br>- Two CSV documents (Extracted99PercentileofDrbDemand.CSV and MatechedLayerandState.CSV). For Extracted99PercentileofDrbDemand.CSV, it contains the UE distribution and the 99-percentile value of the DRB demand in each cell, with the data format of:[UserNumCase1, UserNumCase2, UserNumCase3, UserNumCase4, 99-percentile of DRB demand in n28, 99-percentile of DRB demand in n78-1, 99-percentile of DRB demand in n78-2, 99-percentile of DRB demand in n78-3, 99-percentile of DRB demand in n258]. For MatechedLayerandState.CSV, it contains the UE distribution and the matched sleep state in each cell, with the data format of: [UserNumCase1, UserNumCase2, UserNumCase3, UserNumCase4, LxSy in n28, LxSy in n78-1, LxSy in n78-2, LxSy in n78-3, LxSy in n258]. In these dataset, the UserNumCase1, 2, 3 and 4 are the UE population respectively in n258 cell, n78-2 cell, n78-2 cell, n78-3 cell and n28 cell; the LxSy represents different level of Sleep States we assigned to each cell, L represents the layer, S represents the State, LxSy represents the sleep state of Layer x and Sate y. </p> <p>The dataset is designed for machine learning and telecommunications research.</p> <p>More details of how the dataset is simulated and aplicable scenarios can be found in paper: "A ML-Based Resource Allocation Scheme for Energy Optimi-zation in 5G NR", MDPI sensor, 2025.</p> |
| title | RRC State Diagram 5G-NR UE Traffic Model Dataset (5G-RRC-SD-UE) |
| url | https://doi.org/10.5281/zenodo.16265621 |