The widespread use of global navigation satellite system (GNSS) receiver in mobile devices induces the adoption of effective GNSS-based indoor positioning algorithms exploiting low-cost hardware. In a previous study, we proposed a new architecture for indoor positioning system to estimate the user position by utilizing the pseudoranges from the smartphone-embedded GNSS module. The advantages of such a system are low cost and low requirements in terms of hardware-level modification for end users. However, all end users and most application developers do not have permission to read the pseudoranges from the embedded GNSS modules. Instead of pseudoranges, the user positions are easily obtained from the GNSS module in any mobile device. Thus, we further improve our positioning algorithm based on the position obtained from the embedded GNSS module rather than the pseudoranges. This position does not correspond to the true one since the indoor signal is non-line-of-sight. Thus, it is named the pseudo-position. The key to the improved algorithm is that the distances from the user terminal to the indoor transmitting antennas are calculated using the differences between the position of the outside antenna and the pseudo-position. The algorithm is tested using a simulated GNSS-based indoor positioning system which is implemented on a GNSS software receiver. The simulation results show that the indoor positioning system is able to provide horizontal positioning with meter-level accuracy in both static and dynamic situations. Additionally, the proposed method improves the robustness of the indoor positioning system to the non-synchronization measurements.
Read full abstract