Spectrum Challange 2 Dataset

该数据集已获准公开发布，无限制分发。数据集包含在两个文件中 - scrimmage4_link_dataset.pickle 和 scrimmage5_link_dataset.pickle 泡菜文件存储为元组列表，每个列表对应一个链接，并包含两个元素。每个元素的长度等于该链接中的帧数 - 它因链接而异。第一个元组包含参数 - 1. 信噪比 ('snr') - 1 个元素 2. 调制和编码方案 ('mcs') - 1 个元素 3. 链路的中心频率 ('centerFreq') - 1 个元素 4. 链路的带宽 ('bandwidth') - 1 个元素 5. 功率谱密度 ('psd') - 16 个元素 因此，链路的第一个元组的每个元素的总宽度为 20。第二个元组包含传输成功（'rxSuccess'）。如果为1，则没有帧错误，如果为0，则有帧错误。以下是代码中提到的数据集文件的链接（每个混战一个 pickle 文件）： 混战 4 (547.5 MB) 镜像 混战 5 (979.7 MB) 镜像 包含有关每个比赛的完整信息的更大数据集也可用。有关完整数据集结构的更多详细信息，请参阅 SC2_Dataset_Documentation.pdf。 SC2_Dataset_Technical_Design_Report.pdf 包含有关数据集采集过程的更多信息。这是完整数据集的链接（每个匹配的单独 sqlite 文件）：完整数据集（135.517 GB）镜像（需要访问请求）请使用以下引文来引用数据集：A. S. M. M. Jameel、A. P. Mohamed、X. Zhang 和A. El Gamal，“Deep Learning for Frame Error Prediction using a DARPA Spectrum Collaboration Challenge (SC2) Dataset”，IEEE 网络快报，doi：10.1109/LNET.2021.3096813。

This dataset has been approved for public release with unlimited distribution. The dataset is contained in two files: scrimmage4_link_dataset.pickle and scrimmage5_link_dataset.pickle. Each pickle file stores a list of tuples, where each tuple corresponds to a communication link and contains two elements. The length of each element equals the number of frames in that link, which varies across different links. The first tuple contains the following parameters: 1. Signal-to-Noise Ratio ('snr') - 1 element; 2. Modulation and Coding Scheme ('mcs') - 1 element; 3. Center frequency of the link ('centerFreq') - 1 element; 4. Bandwidth of the link ('bandwidth') - 1 element; 5. Power Spectral Density ('psd') - 16 elements. Thus, the total number of elements in each entry of the first tuple for a link is 20. The second tuple contains the transmission success status ('rxSuccess'). A value of 1 indicates no frame errors, while a value of 0 indicates frame errors. Below are the download links for the dataset files mentioned in the code (one pickle file per scrimmage): Scrimmage 4 (547.5 MB) [mirror], Scrimmage 5 (979.7 MB) [mirror]. A larger dataset containing complete information for each match is also available. For more details on the structure of the full dataset, please refer to SC2_Dataset_Documentation.pdf. SC2_Dataset_Technical_Design_Report.pdf contains additional information about the dataset acquisition process. Here is the download link for the full dataset (separate sqlite files for each match): Full Dataset (135.517 GB) [mirror] (access request required). Please cite the dataset using the following reference: A. S. M. M. Jameel, A. P. Mohamed, X. Zhang, and A. El Gamal, "Deep Learning for Frame Error Prediction using a DARPA Spectrum Collaboration Challenge (SC2) Dataset", IEEE Networking Letters, doi: 10.1109/LNET.2021.3096813.

搜集汇总

背景与挑战

背景概述

Spectrum Challange 2 Dataset是一个由普渡大学于2021年发布的公开时间序列数据集，专注于无线通信领域的帧错误预测。数据集包含两个混战（scrimmage）的pickle文件，每个文件存储为元组列表，其中包含链路的信噪比、调制编码方案、中心频率、带宽和功率谱密度等参数，以及传输成功标志，用于深度学习模型训练和评估。该数据集旨在支持无线边缘学习研究，特别是基于DARPA频谱协作挑战赛（SC2）的数据，相关论文已发表在IEEE网络快报。