Adaptive Topology Control with Link Quality Prediction for Underwater Sensor Networks.

Lili Wang,Fu Xiao,Cheng Huang

2019-01-01

Abstract:Recent years have witnessed a growing interest in applications of Underwater Sensor Networks(USNs). However hash underwater environments, uncontrollable node mobility and unreliable communication links make the design of network protocol challenging. In this article, we mainly focus on how to maintain efficient and dynamic topology for unshackled USNs deployed in a typical kind of shallow seashore environment. Based on the the probabilistic graph series model, we propose a novel scheme called Topology Control with Link-quality Prediction (TCLP) that seize the time-evloving nature of USNs. During the prediction process, anchor nodes predict the future link quality by time series prediction model, while sensor nodes utilize the spatial correlation of underwater links. During the topology control process, more energy-efficient topology is constructed. We prove that TCLP has a guaranteed performance, and our simulation results show that TCLP can improve the energy efficiency while maintaining relatively high data integrity.