Abstract: Navigation users will significantly benefit from BDS and GPS positioning fusion in terms of availability, accuracy and reliability. However, for single point positioning, systematic biases between multi-GNSS systems cannot be eliminated completely, thus the accuracy of positioning and navigation is not always improved with the un-difference measurements of multi-GNSS systems. In this paper, an integrated BDS/GPS positioning model with unknown systematic parameters that compensates for systematic bias is proposed. Furthermore, a Bayesian estimation of fusion positioning model is specifically investigated in which the priori information of the additional parameters is taken into account. Real data collected from different areas with different types of receivers are used to verify those new algorithms. The results show that (a) receiver-dependent inter-system biases are quite evident, while the size of the system bias varies with the receiver type; (b) the precision of fusion positioning is improved significantly by introducing additional parameters into the functional model; and (c) Bayesian estimation of fusion positioning model can still obtain ideal position solution when the number of visible satellites is not enough.