Trajectory Design for Timing–Range Bias Separation in GNSS-Referenced C-UAS Radar Testing

GNSS-referenced counter-uncrewed aircraft system (C-UAS) radar evaluation depends on implementation choices and the information supplied by the test trajectory. Twenty-nine tracks from four date groups at one site were analysed, using CWA 18150 and KS W 8100 as methodological context. Detection-event and range definitions had the highest four-metric influence scores, 21.7% and 11.4%; excluding trackability reversed their order. To first order, constant radial velocity prevents separate estimation of a timing offset and a constant range bias. Offsets were estimated by fitting the time-shifted reference trajectory, because known-delay simulations showed first-order estimates biased by several seconds for 10 s offsets on curved trajectories. Offsets were identified on 19 tracks, 18 of them in a local $\pm 5$ s search, and ranged from $-14.9$ to $+18.6$ s. Adding a timing term to track-specific range biases improved the fit on every pooled date, but a single offset per date was not consistently favoured. Residuals were strongly autocorrelated, and block-bootstrap uncertainties exceeded formal standard errors by a median factor of 2.0 to 2.6, depending on block length. Radial-velocity variation, residual scale and observation count are connected to a target timing precision, and a reporting form is proposed that retains evaluation choices, corrected and uncorrected metrics, and offset uncertainty. Physical offset sources remain unresolved.