| The WSN is composed of a large number of static or mobile sensor nodes in a self-organizing and multi-hop manner,which is used to collaboratively perceive,collect,process and transmit information of the perceived objects in the geographic area covering by the network,and is widely used in military,aviation,environment,home,medical,health care and many other fields.In the practical application of WSN,the information data perceived by the nodes is meaningless without the location information.Therefore,the sensor node localization in the entire WSN applications is essential.Research on low energy consumption,high efficiency,high precision sensor node localization methods is a hotspot in WSNs.Aiming at the current problems of node localization in WSNs,this dissertation research deeply from localization algorithm efficiency,localizable nodes percentage,network nodes density,the speed of mobile nodes on node localization problems in static and mobile WSNs and presents four kinds of node localization methods with theoretical and practical value in order to improve the positioning accuracy and enhance the localization algorithm.The specific research content and innovation of this dissertation are as follows:1)A distributed two-phase node localization algorithm based on PSO for two-dimensional static WSN is proposed.This algorithm transforms the node localization problem into the optimization problem.Based on the PSO method,it reduces the initial search space by bounding box method to speed up the convergence.Then,we define an additional cost function and introduce a prejudgment phase and a refinement phase to prevent and correct the flip ambiguity error due to near-collinear anchors which improves the positioning accuracy and the localizable percentage.Simulation results indicate the proposed algorithm can locate more unknown nodes with higher precision in less computing time.2)Two kinds of node localization algorithm for three-dimensional static WSN are proposed.One research is that a distributed two-phase node localization algorithm based on PSO for three-dimensional static WSN is proposed by extending the above two-dimensional localization algorithm to three-dimensional space.The other one is an improved APIS algorithm which reduces the algorithm complexity,improves the positioning accuracy and localizable percentage by using vertex counting instead of cube counting and the two-hop anchor nodes information to remove those invalid vertexes.The effectiveness of above algorithms is validated by the simulation experiments.The results show that different improved algorithm can be selected to locate 3D node according to the positioning requirements of different scenes.3)A MCB-based node localization algorithm for mobile WSN is proposed.In order to solve the problem of low sampling efficiency and low positioning accuracy,this algorithm firstly utilizes RSSI ranging technique to build a sample box and adds a preset error coefficient in sampling and filtering phase to increase the success rate of sampling and accuracy of valid samples,then introduces sPSO algorithm to generate new samples and avoid constantly repeated sampling and filtering process.Moreover,the segment dynamic weight for two-hop anchors in objective function makes the sample generation process more accurately and quickly.Simulation results denote this algorithm can achieve higher positioning accuracy under different parameter settings.4)The influence of different node mobility models on the performance of localization algorithm is studied.The node mobility models are the basis of simulation experiments.This dissertation uses MCB algorithm as the simulation object,reasech on the relationship of four classical node mobility models and the performance of localization algorithm in the mobile WSN and point out some problems we should consider about when using node mobility model combining with the results of the simulation experiment. |
PDF Full Text Request