In 2017, the new generation satellite-based augmentation system (SBAS) test-bed was initiated by Australia and New Zealand, which supports the dual-frequency multi-constellation (DFMC) positioning with both GPS and Galileo signals. This new SBAS DFMC service allows the elimination of the first-order term of the ionospheric delays, and extends the service area to the entire footprint of the geostationary satellite. In addition to the satellite clock and orbital corrections, the integrity information is also broadcast by the SBAS satellite to users, so that protection levels can be computed to bound the positioning errors with a pre-defined probability of hazardous misleading information. Different from the aeronautical applications, the ground-based applications for road transport may suffer from new problems in different measurement environments, e.g. complicated multipath behaviours and frequent filter re-initialisations during positioning in urban areas. A new weighting model allowing different impacts of the elevation angles, the signal-to-noise-ratios and the smoothing time after re-initialisations is proposed and compared with the traditional elevation-dependent weighting model. The model is applied to the carrier-smoothed code measurements in different environments, i.e., the open-sky scenario, the suburban scenario and the urban scenario. It is found that the new weighting model effectively de-weights the large residuals in the suburban and the urban scenarios, where the mean values and the standard deviations of the overbounding excess-mass cumulative density function can be significantly reduced for the combined weighted noise and multipath. Using 1 Hz GNSS observations measured in these three measurement environments, the horizontal positioning errors (HPEs) and the horizontal protection levels (HPLs) are computed for different filter smoothing windows. Applying the new weighting model, significant reduction can be observed in the mean HPLs in the suburban and urban scenarios. Among them, the reduction in the HPLs have reached about 35–40% in the suburban scenario. The mean absolute HPEs are also reduced by about 10% in the urban scenario. However, when under the open-sky scenario, the traditional elevation-dependent weighting model is sufficient for the positioning and integrity monitoring using the SBAS DFMC service.