valentinbarral/uwb_los_classification_pdp_dl: v0.1

This first release provides the complete experimental framework used in the paper "Using the Power Delay Profile to accelerate the training of neural network-based classifiers for the identification of LOS and NLOS UWB propagation conditions". The repository includes all the necessary code to reproduce the proposed methodology, covering both: MATLAB-based preprocessing, used to parse raw UWB measurements, extract the Channel Impulse Response (CIR) features, and generate Power Delay Profile (PDP) representations. Python + TensorFlow implementations, used to train and evaluate neural network classifiers for LOS/NLOS identification. Main features of this release: End-to-end reproducibility of the experiments described in the paper. MATLAB scripts for dataset parsing, feature extraction, and generation of randomized training/testing sets. Python scripts for training, testing, and evaluating LOS/NLOS classifiers using PDP-based inputs. Result generation and visualization tools for performance analysis. This release is intended as a baseline and reference implementation for researchers working on UWB propagation analysis, LOS/NLOS classification, and machine learning-based channel characterization.