An efficient solution of real-time data processing for multi-GNSS network

摘要: Global navigation satellite systems (GNSS) are acting as an indispensable tool for geodetic research and global monitoring of the Earth, they have been rapidly developed over past few years with abundant GNSS networks, modern constellations, significant improvement in mathematic models data processing. However, due to increasing number satellites stations, computational efficiency becomes a key issue it could hamper further development applications. In this contribution, problem is overcome from aspects both dense linear algebra algorithms processing strategy. First, order fully explore power microprocessors, square root information filter solution based on blocked QR factorization employing many matrix–matrix operations possible introduced. addition, algorithm complexity decreased by centralizing carrier-phase observations ambiguity parameters, well performing real-time resolution elimination. Based simulated matrix, we can conclude that compared unblocked factorization, greatly improve magnitude nearly two orders personal computer four 3.30 GHz cores. Then, 82 globally distributed validated multi-GNSS (GPS/BDS/Galileo) clock estimation. The results suggest will take about 31.38 s per epoch method. While, without any loss accuracy, only takes 0.50 0.31 s our new float fixed solutions, respectively.