Consensus Estimation of Flow Field via Underwater Sensor Networks in Inhomogeneous Medium

Jianhang Zhou, Jing Yan, Xian Yang, Xiaoyuan Luo, Cailian Chen, Xinping Guan

Abstract: This brief studies the flow field estimation issue with underwater sensor networks (USNs) in inhomogeneous medium. We first adopt the ray tracing model to compensate the range bias caused by straight-line propagation, through which a triangle-division-based transmission protocol is designed to collect the travel time differences between any two sensor nodes. With the collected time differences, a flow field estimation problem with acoustic tomography is constructed for USNs. To solve the above problem, a consensus-based unscented Kalman filter (UKF) is designed to minimize the sum of all local measurement errors of sensor nodes. Of note, the triangle-division-based transmission protocol in this brief can avoid the bias under the stratification effect, and meanwhile, the consensus-based estimator can improve the estimation accuracy. Finally, the effectiveness of our solution is verified by simulation and experimental results.