GNSS positioning in deep urban city with 3D map and double reflection

Accurate 3D city map becomes available and commercialized for various applications. 3D map information could improve the performance of GNSS positioning in the city urban environment by recognizing the Non-Line-Of-Sight and multipath signal. In ENC 2015, we have presented the idea of the position hypothesis based GNSS positioning method using 3D maps, called 3D-GNSS. After that, we further improved the 3D-GNSS with considering the differential correction, which has been published in ENC 2016. In our previous works, 3D-GNSS corrects the pseudorange delay by recognizing the single reflection path of satellite signals using ray-tracing and 3D map information. However, the double reflection, which means that the signal is reflected twice from satellite to receiver, is quite possible to be existing, especially in deep urban area with many skyscrapers. Thus, this paper proposes to improve the positioning performance by considering both single and double reflections in our 3D-GNSS method. The result shows that the positioning error could be reduced by 0.8 meter with the proposed idea.