For different ambiguity products, this paper introduced the principle and methodology of the PPP-AR methods based on the dual-frequency ionosphere-free combination model, made detailed analysis of the experimental results in terms of four indexes: positioning accuracy, convergence time, the time to first fix and ambiguity fix rate. The results show that: the ambiguity fix solutions of phase bias product and fractional cycle bias product for the single GPS system and single Galileo system experimental schemes are 0.53 cm, 0.54 cm, 1.05 cm and 0.59 cm, 0.58 cm, 1.31 cm, respectively, which are 49.0%, 32.5%, 15.3% and 41.6%, 20.5%, 3.7 % improved compared with the floating solution. The phase bias product is able to improve the convergence time of the single GPS system by 22.9 % from 17.5 min to 13.5 min, but the ambiguity product improves the convergence time of the Galileo system by less than 10 %, the time to first fix was around 21 min and 27 min for the single GPS system and the single Galileo system, respectively. The ambiguity fix rate of each PPP-AR products is above 90 % for the single GPS system, while the ambiguity fix rate of the single Galileo system is around 85 %.. With the addition of the BDS system, the positioning accuracies of the multi-system phase bias products reach 0.45 cm, 0.49 cm, and 1.02 cm, which are improved by 32 %, 34.7 %, 17.1 %, the convergence time is shortened to less than 10 min, and the time to first fix is reduced to around 10 min, and all the ambiguity products can maintain ambiguity fix rate more than 95 %.