An Accurate Clock Synchronization Method among Moving Objects Based on Maximum Likelihood Estimation

We consider a problem of accurate clock synchronization among multiple moving objects in a three-dimensional (3D) region. To estimate clock offset between two objects, One Way Ranging and Two Way Ranging methods have been commonly applied, but in this paper, we first reveal that they do not perform accurately at all when they are in motion. Next, to solve the problem, we propose a clock offset estimation method among two objects, where each object broadcasts not only its own clock information but also its own location information. The proposed clock offset estimation method is based on Maximum Likelihood (ML) estimation taking into consideration that the measured distance between them is not Gaussian- but 3D-Rician-distributed. Then, we propose a clock synchronization method among multiple objects using the ML-based clock offset estimation results, which is Based on a consensus algorithm. In computer simulation, we show that the proposed clock synchronization method is effective, achieving the clock synchronization accuracy of less than ten nanoseconds among ten moving objects, accuracy of which is required for future wireless communication tools among moving cellular base stations.