Network-based Geometry-Free Three Carrier Ambiguity Resolution and Phase Bias Calibration

Continuously operating reference stations (CORS) are increasingly used to deliver real-time and nearreal-time precise positioning services on a regional basis. A CORS network-based data processing system depends on two types of measurements: (1) ambiguity-resolved double-differenced (DD) phase measurements, and (2) phase bias calibrated zero-differenced (ZD) phase measurements. This paper describes generalized, network-based geometry-free models for three carrier ambiguity resolution (TCAR) and phase bias estimation for DD and ZD code and phase measurements. Firstly, the geometry-free TCAR models are constructed with two Extra-Widelane (EWL) and/or Widelane (WL) virtual signals to allow for rapid ambiguity resolution (AR) for DD phase measurements without distance constraints. With an ambiguity-resolved WL phase measurement and the ionospheric estimate derived from the two EWL signals, an additional geometry-free equation is formed for the third virtual signal independent of the previous two. AR with the third geometry-free model requires a longer period for averaging than the first two, but is also distance-independent. A more general formulation of the geometry-free model for a baseline or network is also introduced, where all the DD ambiguities can be more rigorously resolved using the LAMBDA method. Secondly, the geometry-free models for calibration of three carrier phase biases of ZD phase measurements are similarly defined for selected virtual signals. A network adjustment procedure is then used to improve the ZD phase biases with known DD integer constraints. Numerical results from experiments with 24-hour dual-frequency GPS data from three US CORS stations baseline lengths of 21km, 56km and 74km confirm the theoretical predictions concerning AR reliability of the network-based geometry-free algorithms.