Global navigation satellite system (GNSS) is the most used outdoor positioning system that utilizes high altitude and lower dynamics of the geostationary Earth orbit (GEO) and medium Earth orbit (MEO) satellites. Due to its lower signal strength and lower Doppler shift, it suffers from multipath and NLOS signals from the urban canyon environments. Unlike conventional GNSS, the dominating feature of low Earth orbit (LEO) satellites: the high Doppler shift frequency, large signal strength, and the available LEO constellations, make it a potential alternate candidate for conventional GNSS to address the LEO satellite-based positioning, navigation, and timing (PNT) system. In signals of opportunity (SOPs)-based positioning utilizing LEO satellites, ephemeris data is calculated using two-line element (TLE) files that induce error over time. To handle the erroneous measurement, an additional base receiver with a known position can be used to compensate for the effect of ephemeris error while estimating the position of the user terminal (UT). However, the Doppler shift measurement error due to the ephemeris error is not compensated for the long baseline (the distance between the base receiver and UT), which degrades the positioning accuracy. Moreover, the lack of clock synchronization between the base receiver and UT induces erroneous Doppler shift measurements. Motivated by this, a double-difference Doppler shift-based precise positioning framework is proposed, coined 3DPose, to handle the clock synchronization issue between the base receiver and UT, and positioning degradation due to the long baseline. The proposed 3DPose framework utilizes double-difference Doppler shift measurements to eliminate the clock synchronization issue, along with a novel ephemeris error correction algorithm to improve UT positioning accuracy in case of the long baseline. To this end, the Doppler shift measurement error due to the erroneous ephemeris data is characterized and corrected, considering the position error of satellites in the tangential direction using the ephemeris error correction algorithm. To validate the proposed framework, the positioning outcomes of the proposed framework are compared with the existing differential Doppler positioning method regarding positioning error in three different scenarios. The comparison results confirm that the proposed 3DPose framework demonstrates superior and robust positioning accuracy compared to the benchmark algorithm.
