RadioBench

RadioBench是一个由贝尔实验室创建的数据集，包含5个工业级射频定位数据集，用于研究和基准测试学习型定位网络在实际环境中的部署能力。这些数据集涵盖了不同的环境特性，旨在解决射频信号分布变化对学习型定位系统的影响。数据集的创建过程涉及在不同环境中进行射频测量，以捕捉环境依赖的传播条件。RadioBench的应用领域包括无线定位系统的性能评估和优化，特别是在面对分布变化时的鲁棒性分析。

RadioBench is a dataset created by Bell Laboratories, which contains 5 industrial-grade radio frequency (RF) positioning datasets. It is designed for research and benchmarking of the deployment capability of learned positioning networks in real-world environments. These datasets cover diverse environmental characteristics, aiming to investigate the impact of variations in radio frequency signal distributions on learned positioning systems. The development of RadioBench involved conducting RF measurements across various environments to capture environment-dependent propagation conditions. Its application fields include performance evaluation and optimization of wireless positioning systems, especially robustness analysis under distribution shift scenarios.