Research On Localization Techniques Of Mobile Node In Underwater Sensor Networks

Underwater sensor networks(UWSN)have aroused increasing interests in the realm of underwater acoustic research,because of its potential in a wide variety of civil and military applications,such as marine observation,oceanic exploration,maritime national defense and security,and so on.For UWSN,the location information of the nodes is critical and a prerequisite for many applications.In ocean environment,UWSN nodes are not static,this is the reason why different anchor nodes'(position known node)locator beacons can not reach the unknown node at the same time.Therefore,the traditional node localization methods based on synchronization measurements can't work in UWSN.In this paper,the localization methods of mobile nodes in UWSN were studied.The main works of this paper are as follows:(1)Since the marine environment noise does not follow the Gaussian distribution,the alpha steady-state distribution is used to model the marine environment noise.Then,a maximum likelihood DOA estimation method based on improved cuckoo search algorithm is proposed,which improves the accuracy of DO A estimation.(2)The water medium is inhomogeneous and the sound speed varies depending on several parameters,e.g.,the temperature,pressure and salinity.As a result,sound waves do not necessarily travel in straight lines.Ignoring this stratification effect could lead to considerable bias in the range estimates.This paper proposes a TOA method based on the stratification effect compensation.Assume that the velocity profile is only vertically stratified and the latest localization results of unknown node are known.By tracking the actual transmission path of acoustic wave,we can get accurate distance between unknown node and anchor nodes.(3)This paper presents a localization method based on maximum likelihood algorithm to solve tracking problem.In this method,the measurements at different time in the positioning period is converted into the measurements at the initial time of the positioning period by taking advantage of the speed of the unknown node,and then the maximum likelihood algorithm is used to estimate the motion trajectory of the unknown node.The Crame-Rao lower bound is an important indicator of the parameter estimation performance,and this paper deduces and analyzes the Crame-Rao lower bound of this localization method.The simulation results show that this method has higher localization accuracy than the traditional one.And the performance of this method can reach its Crame-Rao lower bound,so it is highly effective.(4)A new likelihood function is deduced from the principle of the vector space differential transformation.On this basis,a new localization method based on maximum likelihood algorithm can be proposed.In this method,the measurements at different time in the positioning period are converted into the measurements at the initial time of the positioning period by taking advantage of the speed of the unknown node,and then the new likelihood function is used to estimate the motion trajectory of the unknown node.Similarly,this paper deduces and analyzes the Crame-Rao lower bound of this localization method.The simulation results show that this method has higher localization accuracy than the traditional one.