5G-SliciNdd

This dataset repository comprises four distinct files (in arff and csv format), including the "Global" dataset, representing the combined 5G network traffic, and three individual datasets resulting from the segmentation of "Global" based on slice type (eMBB, mMTC, URLLC). These subsets offercomprehensive resources for researchers and professionals interested in network security and intrusion detection within the complex landscape of 5G networks. These datasets empower you to advance your research, develop effective intrusion detection models, and delve into the unique security challenges associated with diverse 5G network traffic profiles.This dataset is the result of an extensive research effort, primarily based on the work of Samarakoon et al., which is outlined in their reference:*Reference:<br>Samarakoon, Sehan, Yushan Siriwardhana, Pawani Porambage, Madhusanka Liyanage, Sang-Yoon Chang, Jinoh Kim, Jonghyun Kim, and Mika Ylianttila. "5G-NIDD: A Comprehensive Network Intrusion Detection Dataset Generated over 5G Wireless Network." IEEE Dataport (December 2, 2022). doi: https://dx.doi.org/10.21227/xtep-hv36.*To produce the 5G-NIDD datasets, we embarked on a meticulous process that involved enriching the base dataset through various steps:Obtaining 5G-NIDD: We acquired the 5G-NIDD dataset, created by Samarakoon et al., which served as our foundation.Global Dataset: The "Global" dataset was created to represent the combined 5G network traffic. It includes a variety of traffic types, showcasing the diversity of the 5G environment.Slice Segmentation: The "Global" dataset was then split based on slice type, leading to the creation of three distinct subsets. These subsets are:eMBB Subset: Focusing on enhanced Mobile Broadband traffic.<br>mMTC Subset: Concentrating on massive Machine-Type Communication traffic.<br>URLLC Subset: Targeting Ultra-Reliable Low-Latency Communication traffic.This segmentation enabled us to examine and analyze intrusion detection in the context of these specific 5G traffic profiles, highlighting the unique security challenges that arise within each slice.本数据集仓库包含4个格式为arff与csv的独立文件，分别为表征合并后5G网络流量的"Global"数据集，以及基于切片类型对"Global"数据集进行分割得到的3个独立子集（增强移动宽带（enhanced Mobile Broadband, eMBB）、海量机器类通信（massive Machine-Type Communication, mMTC）、超可靠低延迟通信（Ultra-Reliable Low-Latency Communication, URLLC））。 本数据集可为关注复杂5G网络场景下网络安全与入侵检测方向的研究者与专业人士提供全面的研究资源，助力推进相关研究、构建高效的入侵检测模型，以及深入剖析多样化5G网络流量特征下的独特安全挑战。 本数据集基于Samarakoon等人的研究工作构建，是一项系统性研究的成果，其相关参考文献如下： *参考文献:<br>Samarakoon, Sehan, Yushan Siriwardhana, Pawani Porambage, Madhusanka Liyanage, Sang-Yoon Chang, Jinoh Kim, Jonghyun Kim, Mika Ylianttila. "5G-NIDD：基于5G无线网络生成的全面网络入侵检测数据集". IEEE Dataport (2022年12月2日). doi: https://dx.doi.org/10.21227/xtep-hv36.* 为构建5G-NIDD数据集，我们开展了一系列严谨细致的工作，通过多步骤丰富基础数据集： 1. 获取5G-NIDD：我们获取了Samarakoon等人构建的5G-NIDD数据集，将其作为本数据集的构建基础。 2. "Global"数据集：创建"Global"数据集以表征合并后的5G网络流量，该数据集涵盖多种流量类型，充分展现了5G网络环境的多样性。 3. 切片分割：随后基于网络切片类型对"Global"数据集进行拆分，得到3个独立子集，分别为：<br> - eMBB子集：聚焦增强移动宽带（enhanced Mobile Broadband, eMBB）流量<br> - mMTC子集：专注于海量机器类通信（massive Machine-Type Communication, mMTC）流量<br> - URLLC子集：针对超可靠低延迟通信（Ultra-Reliable Low-Latency Communication, URLLC）流量 该切片分割方式可支持研究者针对这些特定5G流量场景开展入侵检测相关的研究与分析，凸显各网络切片内独特的安全挑战。