YAO HU

Ms. YAO HU


Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
14 November 2025 in Remote Sensing
Verified
Journal Article
Open access
PDF file
ZWZhouji Wang
Zhouji Wang
Daqian LyuDaqian Lyu
Daqian Lyu
PZPeiyuan Zhou
Peiyuan Zhou
YGYulong Ge
Yulong Ge
YHYAO HU
YAO HU
XYXiaoniu Yang
Xiaoniu Yang
Timing and time synchronization are critical capabilities of Global Navigation Satellite Systems (GNSSs), but their performance deteriorates significantly in challenging environments like urban canyons and tunnels. To address this issue, this paper proposes the Distributed Sensor Time Synchronization architecture (DSTS), a novel architecture integrating Bayesian filtering with deep reinforcement learning. DSTS utilizes Bayesian filtering to fuse Time-of-Flight (ToF) measurements with Channel Impulse Response features for real-time compensation of non-linear errors and accurate path state prediction. Concurrently, the Deep Deterministic Policy Gradient (DDPG) algorithm trains each node into an intelligent agent that dynamically learns optimal synchronization weights based on local information like neighbor clock stability and link quality. This allows the architecture to adaptively amplify reliable nodes while mitigating the negative effects of unstable peers and adverse channels, ensuring high accuracy and availability. Simulation experiments based on a real-world UWB dataset demonstrate the architecture’s exceptional performance. The Bayesian filtering module effectively mitigates non-linear errors, reducing the standard deviation of ToF measurements in NLOS scenarios by up to 51.6% (over 41.2% consistently) while achieving high path state prediction accuracy (>85% static, >95% simulated dynamic). In simulated dynamic and heterogeneous networks, the DDPG algorithm achieves a synchronization accuracy better than traditional average-consensus algorithms, ultimately reaching a frequency and phase precision of 4×10−10 and 5×10−10 s, respectively.

19 September 2025 in Measurement Science and Technology
Verified
Journal Article
Closed access
YHYAO HU
YAO HU
Daqian LyuDaqian Lyu
Daqian Lyu
PZPeiyuan Zhou
Peiyuan Zhou
YGYulong Ge
Yulong Ge
JZJie Zhang
Jie Zhang
XCXinyun Cao
Xinyun Cao
Precise point positioning (PPP) is an advanced technique for GNSS-based time transfer, supporting both static and kinematic applications. However, kinematic scenarios pose significant challenges due to environmental sensitivity, which degrades the robustness of clock parameter estimation. Conventional approaches often model receiver clock offsets as white noise, leading to suboptimal performance under dynamic conditions. To address this issue, we propose an advanced clock modeling method for real-time kinematic timing. The model integrates a second-order state transition framework with differential carrier phase-based frequency estimation, aiming to improve clock stability and filtering accuracy. The study is organized into three parts. First, a comparative evaluation of four real-time precise products (CAS, CNES, WHU, and HAS) was conducted across multiple International GNSS Service (IGS) stations to assess their baseline timing performance. Results confirm that HAS demonstrates favorable timing stability, supporting its use in dynamic applications. Second, the conventional model was compared with white noise strategies using simulated kinematic data. The improved model significantly reduced re-convergence effects and improved filtering accuracy, with standard deviation (STD) reductions of 39.63% for CAS and 41.60% for HAS. Third, to validate the advanced model in real-world performance, the model was applied to a field experiment using a moving vehicle and HAS corrections. The results show a 30% improvement in STD and notable enhancements in clock frequency estimation. Modified Allan deviation analysis further confirmed improved short-term and long-term stability, with short-term fluctuations approaching the 10−14 level. These results demonstrate the feasibility and robustness of the proposed clock model for kinematic GNSS timing applications, particularly when integrated with real-time products like HAS.

27 February 2026 in Measurement Science and Technology
Verified
Journal Article
Open access
YHYAO HU
YAO HU
Daqian LyuDaqian Lyu
Daqian Lyu
PZPeiyuan Zhou
Peiyuan Zhou
YGYulong Ge
Yulong Ge
ZWZhouji Wang
Zhouji Wang
RZRangang Zhu
Rangang Zhu
In multi-GNSS positioning, navigation and timing, the inter-system time offset must be handled explicitly. For Galileo–BDS Time Offset (GBTO), PPP-BCE is limited by signal-in-space errors, and fixing GBTO as a prior is fragile due to sub-nanosecond datum and inter-system bias drifts between HAS (GTRF) and PPP-B2b (BDCS). We assess HAS and PPP-B2b and compare three strategies: a PPP-BCE baseline, a post-fit differenced PPP, and a tightly coupled PPP that fuses HAS and PPP-B2b in one EKF and estimates GBTO as a lightly stochastic state. Validations employing 14 d of high-rate analysis at NT05/USUD and an extended two-month experiment at geographically dispersed stations (MIZU, HKSL, and JFNG) demonstrate the method’s long-term robustness. The fused filter gives the lowest short-term scatter: rolling standard deviation 0.1–0.2 ns; the differenced SSR stays below 0.4 ns; PPP-BCE reaches up to 1.0 ns. It also provides the best frequency stability: at NT05, MDEV at τ = 960 s improves five times over PPP-BCE and reaches the 10 −14 level at τ = 10 4 s. Against BIPM, all real-time solutions share a bias of 2–4 ns; the fused method has the tightest day-to-day residuals (NT05 IQR 1.1 ns). Broadcast cross-checks show a time-varying difference near 20 ns between BDGA and the UTC bridge. Estimating GBTO within one HAS and PPP-B2b PPP filter enables autonomous multi-GNSS timing with high temporal resolution, independent of broadcast parameter latency.

22 May 2026 in Measurement Science and Technology
Verified
Journal Article
Open access
Daqian LyuDaqian Lyu
Daqian Lyu
YHYAO HU
YAO HU
PZPeiyuan Zhou
Peiyuan Zhou
YGYulong Ge
Yulong Ge
JZJie Zhang
Jie Zhang
ZWZhouji Wang
Zhouji Wang
The Quasi-Zenith Satellite System Multi-Global Navigation Satellite Systems Advanced Orbit and Clock Augmentation-Precise Point Positioning (MADOCA-PPP) service and Europe’s Galileo High Accuracy Service (HAS) are two freely accessible real-time augmentation services that provide orbit and clock corrections for PPP. This paper presents an evaluation of these two services for precise time transfer. Archived corrections and GNSS observations collected from May 10 to 20, 2024, were processed to assess (i) orbit and clock correction quality and (ii) PPP timing performance under static and kinematic conditions. For MADOCA-PPP, the mean orbit differences in the radial ( R ), along-track ( A ), and cross-track ( C ) directions were 0.033/0.041/0.025 m for GPS, 0.030/0.072/0.047 m for GLONASS, and 0.042/0.056/0.031 m for Galileo. In terms of the standard deviation of the clock deviation, the values of GPS and Galileo systems are smaller than that of the GLONASS system. For the HAS system, the orbital deviation and clock deviation sequences show higher noise levels, and the Galileo orbital deviation shows significant systemic fluctuations during the evaluation period. In the PPP time transmission experiment, the timing statistics obtained by the solution scheme based on MADOCA-PPP correction are close to the final product of Center for Orbit Determination in Europe, while the solution scheme based on HAS shows greater dispersion and decreased frequency stability over the medium-long averaging period, and the sensitivity increases in dynamic tests. Overall, this study analyzes the correction behavior and terminal PPP timing performance of MADOCA-PPP and HAS, which provides a practical reference for real-time precision positioning, navigation and timing applications.