大规模MIMO超维信道模型数据集

本数据基于实测信道数据，包括所训练的深度学习信道频域外推器在各实测信道样本上的归一化均方误差。采集时间与地点为2025年1月，清华。原始公开信道数据信道测量平台为ArgosV2platform，基站天线数为54，测量频率为2.4GHz，带宽为20MHz，子载波数为52。原始信道读取通过python。该数据集可支持实测信道映射精度计算，数据集大小为222.03KB

This dataset is built upon measured channel data, and contains the normalized mean square error (NMSE) of the trained deep learning-based channel frequency-domain extrapolator over each measured channel sample. Data collection was carried out in January 2025 at Tsinghua University. The channel measurement platform for the original publicly available channel data is the ArgosV2 platform. The base station is equipped with 54 antennas, the measurement frequency is 2.4 GHz, the channel bandwidth is 20 MHz, and the number of subcarriers is 52. Raw channel data was read using Python. This dataset enables the calculation of measured channel mapping accuracy, with a total size of 222.03 KB.

搜集汇总

背景与挑战

背景概述

该数据集基于2025年1月在清华大学实测的大规模MIMO信道数据，包含深度学习信道频域外推器的归一化均方误差，用于评估信道映射精度。数据集规模为222.03KB，覆盖54天线基站、2.4GHz频率和20MHz带宽的测量参数，适用于大规模MIMO无线通信研究。