Factor Graph-Based Spatial-Temporal-Enhanced Attention Network-Assisted TDCP Positioning

Recently, the time-difference carrier phase(TDCP) positioning method based on factor graph has become a promising solution to address the positioning challenges in urban areas, allowing for high-precision results without resolving integer ambiguities. However, in challenging environments such as urban canyons, cycle slips in carrier phase are inevitable, leading to significant degradation in positioning accuracy. To address this issue, this paper proposes a spatio-temporal attention network-assisted TDCP positioning method based on factor graphs. Specifically, to accurately evaluate the quality of satellite signals, we fully exploit their spatio-temporal characteristics and introduce a spatio-temporal attention mechanism into the satellite signal quality evaluation network. Secondly, a deep learning-assisted cycle slip detection method is proposed, which takes into account both network evaluation results and traditional Doppler shift-based cycle slip detection methods, enabling reliable detection of carrier phase cycle slips. Finally, the positioning result is obtained through sliding window optimization in the factor graph. We validate the effectiveness of the algorithm using UrbanNav dataset.