Mitigation of Diffraction Effect in GNSS Positioning Considering Azimuth and PDOP in Canyon Environment

Global navigation satellite system (GNSS) monitoring stations established in canyon environments inevitably face challenges such as multipath effects, diffracted signals, and non-line-of-sight (NLOS) reception due to obstructions. To address this problem, existing methods employ a stochastic model based on GNSS receiver signal-to-noise ratio (SNR) to detect NLOS reception. However, the method only accounts for NLOS reception concerning elevation angles of satellites, neglecting NLOS reception from obstructions at side edges. To overcome this limitation, this paper proposes a threshold of azimuth angle. When the threshold is considered, it does not only improve NLOS detection, but also increases the position dilution of precision (PDOP) value, potentially affecting positioning precision. Thus, we introduced a PDOP stochastic model and proposed a method that integrates both the azimuth angle threshold and PDOP considerations. Experimental results show that the precision of the stochastic model based on equivalent elevation angles improves by approximately 24.25% after considering the threshold of azimuth angle. Furthermore, the precision increases by an additional 8.69% when the PDOP stochastic model is also taken into account.