Abstract: Fast and reliable ambiguity resolution plays a critical role in GNSS real-time precise positioning, however, the computational complexity of finding the optimal integer ambiguity vector directly will be high, because there is strong correction between ambiguities. Thus the lattice theory was introduced in this paper to reduce the correlation for improving the computational efficiency of ambiguity resolution, and the main decorrelation methods currently existed were also analyzed and transformed into the methods of lattice reduction. Then the Bootstrapping success rate was suggested and tested by an experiment to show the performance of these decorrelation methods used for long-baseline ambiguity resolution with multi-frequency GNSS signals.