Online Resource 1 - Real GNSS Data Experiment Dataset for "Hybrid Fourier–LSTM Framework for Adaptive GNSS Anomaly Detection via Multi-Scale and Type-Specific Fusion"

This repository corresponds to Online Resource 1 of the manuscript “Hybrid Fourier–LSTM Framework for Adaptive GNSS Anomaly Detection via Multi-Scale and Type-Specific Fusion.”It contains the reproducibility package for the real GNSS data experiments, including Python scripts, input examples, and output files. Contents: hybrid_gnss_real_data.py — main script for processing real GNSS datasets. .xlsx files — station-wise GNSS position time series (E, N, U). .npz, .json, .csv — model outputs, metrics, and summaries. README.md — detailed instructions for environment setup, input format, and execution steps. Environment:Python ≥ 3.9 with the following dependencies: numpy, pandas, torch, openpyxl. Execution:Run the pipeline with: python hybrid_gnss_real_data.py Outputs include:predictions.npz, scores.npz, site_summary.csv, summary.json, and summary_all.json, which correspond to the results reported in the manuscript.