ESTIMATION OF HARDWARE BIASES AND THEIR IMPACT ON GNSS POSITIONING: A CASE STUDY OF DIFFERENTIAL CODE BIAS

ABSTRACT This study estimates hardware biases, specifically Differential Code Bias (DCB), and evaluates their impact on Global Navigation Satellite System (GNSS) positioning using GPS data from two stations, ABUZ (Zaria) and CGGT (Toro), over DOY 001–003, 2011. The model formulation employs geometry-free linear combinations of pseudo-range code observations to isolate ionospheric effects and biases. Cycle slip detection is performed using the Melbourne-Wübbena combination, with Hatch smoothing applied to reduce noise. DCB estimation using least-squares adjustment with a zero-mean constraint on satellite biases, incorporating Vertical Total Electron Content (VTEC) from CODE IONEX maps converted to Slant TEC via a single-layer mapping function at 506 km shell height. Results show stable receiver DCBs, with ABUZ constant at an average of 1.45 ns (standard deviation 0.07 ns, range 1.38–1.52 ns) and CGGT at -0.62 ns (standard deviation 0.06 ns, range -0.68 to -0.55 ns) across DOY 001–003. Satellite DCBs range from -7.85 to 5.25 ns (average standard deviation 0.10 ns). VTEC averages 28.8 TECU for ABUZ and 28.1 TECU for CGGT. Validation against CODE IGS Repro3 Bias-SINEX products yields a mean bias of 0.00 ns, RMSE of 0.06 ns, and correlation of 0.999 for satellite DCBs, confirming the model accuracy. Uncorrected DCBs may introduce 0.3–0.5 m errors in single-frequency positioning or 5–10 cm in PPP. The baseline methodology provides a practical MATLAB framework for local bias correction, advancing GNSS accuracy in Nigeria