RULE-BASED RECONSTRUCTION AND EXTENSION OF A 5G NETWORK SLICING DATASET WITH V2X INTEGRATION

The rapid expansion of 5G services and the emergence of Vehicle-to-Everything (V2X) applications have created a demand for datasets that accurately represent diverse network slice characteristics. However, most existing 5G network slicing datasets primarily emphasize conventional slice types, including enhanced Mobile Broadband (eMBB), Ultra-Reliable Low-Latency Communications (URLLC), and massive Machine-Type Communications (mMTC), while V2X remains insufficiently represented, thereby limiting the development of data-driven models for intelligent network management and vehicular communication systems. This study introduces a hybrid framework that integrates data-driven analysis with rule-based modeling to extend an existing 5G network slicing dataset. The framework statistically extracts attribute characteristics of conventional slices and incorporates V2X attributes derived from recent literature through a structured reclassification approach. Statistical techniques are applied to determine attribute value ranges for canonical slices, while V2X parameters are synthesized from documented communication requirements. Slice-specific data instances are then generated and integrated into a unified dataset covering four slice categories. The extended dataset contains 3,000 instances distributed across eMBB, mMTC, URLLC, and V2X slice types, with clearly differentiated attribute distributions aligned with their respective Quality of Service requirements. The rule-based reclassification method ensures consistent, interpretable, and reproducible slice labeling, validating the structural integrity of the reconstructed dataset. Overall, the proposed framework provides a scalable and reproducible foundation for advancing machine learning-based slice classification and intelligent resource management in 5G networks.