Abstract: Proposed a mobility prediction in underwater sensor networks is based on extensible positioning method (SLMP).In SLMP,positioning uses hierarchical implementation,the localization process is divided into two parts:anchor node localization and ordinary node localization.In the process of locating,each node is to predict its future mobility model,location information is based on its past known,the predicted flow patterns can be estimated based on its future position.Anchor nodes with known locations of the network will control the positioning process,in order to balance trade-offs between accuracy,coverage,and the communication cost.We conduct extensive simulation;results show that,SLMP can greatly reduce the communication cost while maintaining positioning,positioning relatively high coverage rate and positioning accuracy.