Objectives   The undifferenced ambiguity is recovered to the integer characteristics after the fractional cycle bias (FCB) product correction, which significantly shortens the convergence time of precision point positioning (PPP). When the uncombined FCB products are estimated, the original L1 and L2 FCB cannot be accurately separated due to the limitation of global ionospheric model accuracy.On one hand, due to the limitation of the accuracy of global ionospheric grid model, L1 and L2 ambiguity cannot be separated accurately in real-time, which makes the inconsistent for user end using the uncombined FCB products. On the other hand, due to the difference between the ionospheric combined FCB products and the uncombined FCB products, the user end with uncombined PPP model cannot use the ionospheric FCB products.

  Methods   A partial ambiguity resolution method for uncombined PPP using the ionosphere-free combined FCB product is proposed, which considers the consistency between the algorithm of generating the ionosphere-free combined at server end FCB product and the algorithm of the uncombined ambiguity resolution at user end. Constructing the narrow-lane ambiguity by using the raw ambiguity and wide-lane ambiguity of single difference between satellites, the ionosphere-free combined FCB product is used to fix the ambiguity step by step. Consisting of about 120 global multi-GNSS experiment (MGEX) stations are used to generate the ionosphere-free combined FCB and uncombined FCB products, and 10 stations which are not in the service end are selected for evaluation and validation. For the server end with 120 MGEX stations, 97.3% of the wide-lane ambiguity residuals and 96.8% of the narrow-lane ambiguity residuals are distributed after ionosphere-free combined FCB products correction. 96.7% of the wide-lane ambiguity residuals and 97.7% of the narrow-lane ambiguity residuals are distributed after uncombined FCB products correction.

  Results   The experimental results show that the positioning accuracy of the proposed method is improved by 25.0% and the convergence time is shortened by 21.1% in static condition, 26.7% and 17.9% in dynamic condition, respectively.

  Conclusions   Compared with the traditional FCB method, the proposed method can improve the positioning accuracy and shorten the convergence time, which can further broaden the application scenarios of PPP.