Content area
A novel DSTS architecture was proposed, integrating Bayesian filtering with DDPG reinforcement learning to solve synchronization in Challenging environments. The DSTS architecture achieved a final frequency synchronization precision of
The fusion of ToF and CIR data via Bayesian filtering provides an effective method to address non-linear communication errors and propagation path state uncertainty. The use of a DDPG agent as an “attention-like” mechanism is a viable strategy for managing network heterogeneity. 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
Details
Accuracy;
Intelligent agents;
Protocol;
Algorithms;
Clocks & watches;
Synchronization;
Distributed sensor systems;
Time synchronization;
Machine learning;
Localization;
Reinforcement;
Heterogeneity;
Frequency synchronization;
Propagation;
Remote sensing;
Bayesian analysis;
Canyons;
Sensors;
Design;
Crystal oscillators;
Error reduction;
Impulse response;
Deep learning;
Real time;
Street canyons;
Filtration;
Global navigation satellite system
; Lyu Daqian 2
; Zhou, Peiyuan 3 ; Ge Yulong 4 ; Hu, Yao 4 ; Zhu Rangang 1 ; Wang, Wei 5 ; Yang Xiaoniu 5 1 College of Electronic Engineering, National University of Defense Technology, Hefei 230037, China; [email protected] (Z.W.);
2 College of Electronic Engineering, National University of Defense Technology, Hefei 230037, China; [email protected] (Z.W.);, National Key Laboratory of Electromagentic Space Security, Jiaxing 314033, China
3 Institute of Geospatial Information, Information Engineering University, Zhengzhou 450001, China
4 College of Marine Science and Engineering, Nanjing Normal University, Nanjing 210023, [email protected] (Y.H.)
5 National Key Laboratory of Electromagentic Space Security, Jiaxing 314033, China