Maritme Dataset - 6G-NTN

Real-world open datasets play a pivotal role in the development of AI models addressing radio control optimization problems. As a matter of fact, acquiring suitable datasets can be arduous. Therefore, this dataset is part of the ones identified in by (https://doi.org/10.1109/FNWF58287.2023.10520488) which pertinent to real-world open datasets representing realistic traffic pattern, network performances and demand for fixed and dynamic user terminals, enabling a variety of uses cases. This maritime dataset comprises of two mat files for each hour. Since each grid contains 100 x100 km haversian distance area, for the sake of generality, the first mat file (Coordinates) contains the coordinates of the central point of each grid. For instance, the central coordinates correspond to the points where the numbers are written in Figure Below. Whereas the second mat file (Density_Flow) is corresponding to these numbers themselves which means the number of vessels present within this 100 x100 km area at given hour. Both .mat files are with the dimensions of 17 x 35 and each cell of one file corresponds to the same cell of other file. The name of the files means xxx_Day_1_time_start_to_time_end_UTC.time_start_to_time _end means that the file contain data from start hour to end hour For Examplexxx_Day_1_00_to_01_UTCmeansthis file belongs to day 1 (since we have 1 week data) from 12 am to 01 am in UTC time. Each day contains 24 files with 24 hours.面向无线电管控优化问题的人工智能（Artificial Intelligence, AI）模型开发中，真实世界开源数据集发挥着关键支撑作用。然而，获取适配的开源数据集往往颇具挑战。因此，本数据集源自文献[1]（https://doi.org/10.1109/FNWF58287.2023.10520488）中收录的真实世界开源数据集之一，其涵盖贴合实际的交通模式、网络性能以及固定与动态用户终端的需求相关数据，可支撑多种应用场景。 本海事数据集每小时对应两个MATLAB数据文件（.mat）。由于每个网格对应100×100千米的大圆距离（haversian distance）区域，为保证数据的通用性，首个MAT文件（Coordinates）存储各网格中心点的坐标信息。例如，中心坐标即下图中标注数字的对应位置。而第二个MAT文件（Density_Flow）则存储上述标注数字的具体数值，即对应小时内该100×100千米区域内的船舶数量。两个MAT文件的维度均为17×35，且两个文件中每个单元格的位置一一对应。 本数据集文件的命名规则遵循格式：xxx_Day_{天数}_{起始时刻}_to_{结束时刻}_UTC，其中start与end分别代表数据采集的起始与结束小时，即该文件包含从起始小时至结束小时时段内的观测数据。例如，xxx_Day_1_00_to_01_UTC代表该文件属于第1天（本次数据集共包含1周的观测数据），对应协调世界时（UTC）0时至1时的时段。单日共包含24个数据文件，对应全天24个小时。