Outage-resilient real-time precise point positioning with SSR prediction

Real-time precise point positioning (PPP) based on state-space representation (SSR) corrections is widely used in high-accuracy real-time applications, but its performance degrades rapidly when SSR streams are interrupted or valid satellite corrections are temporarily unavailable. To address this problem, this study proposes an outage-resilient PPP strategy that combines three components: an issue of data (IOD) consistent transformation of SSR orbit and clock corrections to remove artificial jumps at broadcast-ephemeris switches, short-term prediction of orbit and clock SSR corrections using polynomial fitting, and adaptive down-weighting of predicted corrections in the PPP filter, followed by a gradual ramp-in when real-time SSR corrections resume. The proposed method was evaluated using GPS, Galileo, and BDS data under simulated short-term SSR outages. The overall statistical analysis was conducted using one week of data from GPS Week 2363, while representative IOD and single-station examples were taken from 20 April 2025. The results show that the proposed strategy improves positioning stability during outages and reduces discontinuities during recovery. For outages shorter than 600 s, the average RMS values in the north, east, and up components are reduced by 48.08%, 31.58%, and 51.44%, respectively, relative to the direct-recovery strategy. During the recovery period, the corresponding reductions reach 49.62%, 40.08%, and 56.52% when state retention and ramp-in are jointly applied. These results demonstrate that the proposed framework provides a practical and effective way to improve the robustness and recovery smoothness of real-time PPP under short SSR interruptions.