GPS-SpoofIQ: Measured and Simulated Raw GPS L1 C/A Spoofing Data for Navigation and Timing Security Research

The simulated component includes three full-channel position-spoofing scenarios with a common target of 600 m eastward displacement and pull-off rates of 6, 12, and 24 m/s; three full-channel timing-spoofing scenarios with a common target receiver clock-bias offset of +2 μs and pull-off rates of 20, 40, and 80 ns/s; and one partial-channel same-PRN superposition scenario in which a spoofing component is injected only into PRN 32 while the other six visible PRNs remain unaffected.

The simulations use a fixed RINEX broadcast ephemeris and retain the authentic signal components, injection ground truth, and labels identifying spoofed and healthy PRNs. All raw signals are sampled at 25 MHz and stored as headerless little-endian signed 16-bit interleaved complex I/Q samples in the sequence I0, Q0, I1, Q1, and so forth. In addition to the raw signals, the repository provides scenario parameters, file manifests, simulated injection ground truth, per-channel observables generated by a scalar software receiver, navigation solutions.

The dataset supports studies of spoofing detection and classification, correlation-peak takeover, receiver tracking robustness, partial-channel anomaly detection, fault detection and exclusion, receiver autonomous integrity monitoring, and cross-scenario evaluation of anti-spoofing methods.