Delay Compensation-Based Time Synchronization Under Random Delays: Algorithm and Experiment

摘要

This article proposes a delay compensation-based time synchronization (DCBTS) algorithm for wireless sensor networks (WSNs) with random bounded communication delays. First, we point out that in the framework of consensus-based time synchronization (CBTS) algorithms, the offset estimate is generally divergent in networks containing at least two root nodes due to immeasurable delays even if the drift estimate converges at a rate no slower than $1/k$ , where $k$ is the discrete time. Second, we introduce a delay compensation mechanism in the offset estimation algorithm and utilize the first difference of the offset estimate to adaptively adjust the delay compensation. An upper bound of the adjusting gain is also provided, which ensures the boundedness of the offset estimate. Besides simulations under different clock, network, and delay settings, a series of physical experiments have been conducted on our autonomously designed test bed for comparing the DCBTS algorithm with some existing time synchronization algorithms, such as Average TimeSynch (ATS), Weighted-Maximum TimeSynch (WMTS), Least-Square estimation TimeSynch (LSTS), and PISync. Both numerical simulations and physical experiments validate the availability, effectiveness, and better performance of this novel time synchronization algorithm.