Research On Mobile Localization Of UASNs Based On NLOS And Area Filtering

With the continuous exploration of the ocean,underwater acoustic sensor networks(UASNs)are applied in many different fields,and have good application prospects.Many applications need to monitor and determine the locations of underwater sensor nodes in real time,such as water pollution monitoring,military operations,navigation and localization.In the process of accurate localization,the nodes are affected by the complex underwater environment,which easily leads to the increase of localization error.In order to reduce the localization error,this thesis deeply analyzes the characteristics of UASNs and proposes the following algorithm.Aiming at the error and robustness of localization algorithms in UASNs,we proposed two monte carlo mobile localization algorithms: circular ring monte carlo localization(CRMCL)and particle swarm optimization for circular ring monte carlo localization(PRMCL).Firstly,CRMCL algorithm used one-hop anchor nodes to construct circular sampling area and ring filter.We obtained reasonable sample number by defining sampling points density.Then,the relationship between the ring parameter and the filtration area was demonstrated.The reasonable ring parameter was obtained through the simulation experiments,and an efficient filter was constructed to reduce the localization error.The samples filtered by CRMCL algorithm were optimized by particle swarm optimization.PRMCL algorithm not only reduced the number of invalid samples,but also enhanced the robustness of the algorithm.Simulation results show that both CRMCL and PRMCL algorithms have lower localization error and better robustness than monte carlo localization algorithm and other improved algorithms without additional hardware.Aiming at the problem of large localization error of mobile nodes in shallow sea non-lineof-sight(NLOS)environment,this thesis proposed interval error identification time of arrival(IEI-TOA)algorithm.Firstly,IEI-TOA algorithm introduced Euler model to predict the locations of the current moment nodes,and measured the distance between nodes in NLOS environment.Then,the NLOS path maximum theorem under single reflection was proposed by using the spatial distribution relationship among anchor nodes,mobile nodes and obstacles.The minimum value of path recognition interval was solved by projection method.Error formulas were defined to accurately identify the NLOS path,and localization errors were reduced by screening out anchor nodes on the path.Finally,Taylor series was introduced to estimate the locations of mobile nodes.The results show that the localization error of IEI-TOA algorithm is smaller than other algorithms when the line-of-sight and NLOS are confused in shallow water.