Learning-Based Ultra-Wideband Indoor Ranging

Precise indoor positioning has become increasingly important with the rise of Internet of things (IoT), robotics, and autonomous vehicles, etc. In this paper, we focus on time of arrival (TOA) based ranging through channel impulse response (CIR) by machine learning algorithms. The CIR data is obtained in extensive experiments using ultra-wideband (UWB) transceivers as the anchor and tag. We collect a dataset of 412, 172 traces of CIR data across 12 indoor line-of-sight (LOS) scenarios. This dataset is used to train and test three machine learning models, including long short-term memory (LSTM), gated recurrent units (GRU), and multi-layer perception (MLP), to predict the range between the anchor and tag. The results demonstrate that LSTM and GRU models outperform traditional methods and the UWB device's built-in ranging algorithm in terms of mean squared error (MSE), root mean squared error (RMSE), and mean absolute error (MAE).