Research And Realization Of High Precision Node Localization Algorithms In Wireless Sensor Networks

In recent years,with the rapid development of electronic and communication technologies,wireless sensor network(WSN)becomes flexible and is deployed conveniently which make it widely used in military field,environment monitoring,municipal detection,disaster prevention,security,household,transportation,manufacturing and medical fields,and others.A typical WSN is composed of many sensor nodes with limited energy,low power and low energy consumptions.The significant characteristics of sensor nodes are limited computing and communication capabilities,difficult to maintain and update,and limited node life.The accurate position information of sensor nodes in WSN is quite important for disaster rescue and target tracking,network routing protocol and communication network,and localization of people and items,etc.It will be meaningless for WSN applications without location information.Therefore,how to obtain the node position accurately has become the basis and premise of WSN applications,which is also a difficult problem.Aiming at the problem of node localization with high accuracy in WSN and the deficiency of existing algorithms,static and dynamic node localizations are studied respectively in this paper for the typical WSN applications,which break through the constraints of accurate location,low power consumption and high efficiency of node localization.There are a certain of theoretical significances and practical application values.The main research contents of this paper are as following:First of all,for the problems of low anchor node deployment and low node density distribution in static node localization,a static node localization model is proposed.Based on the proposed model,an optimal distance estimation method using node sharing mechanism and the corresponding 3D static node localization algorithm that be applied in real scenes are presented.Combining with the real application scenario requirements including the 3D static node distribution and range-free localization scenario,the designed localization model could obtain the shared relay node between any two sensors in multi-hop range by the acquirement of non-complex network information such as network connectivity.The optimal distance can be estimated by the geometric relations and probability distributions between nodes This paper focuses on the 3D static node localization problem with limited node energy,which could satisfy the requirement on low computation and communication complexities.Secondly,aiming at the problem of dynamic node localization under large scale and high dynamic conditions,a 3D node motion model and accurate representation method of motion state are constructed.A spatial node random motion model and particle swarm optimization(PSO)mechanism are proposed,as well as a hybrid mobile node localization algorithm based on Monte Carlo and improved PSO.An adaptive anchor node selection operator is established by designing a random waypoint motion model,and the objective function is also redesigned in accordance with the 3D dynamic environment.Based on the proposed model and optimization mechanism,the problem of high-precision and real-time localization of 3D mobile nodes is solved in highly random and complex WSN.Combining with the advantages of Monte Carlo algorithm and PSO,the proposed method in this paper overcomes the shortcomings of the mentioned two algorithms,and improves the precision and efficiency of mobile node localization effectively.Finally,in order to build the hardware platform for 3D node localization algorithm under static and dynamic WSN conditions and better realize the algorithm testing,validation and analysis,the overall architecture of node localization prototype system is designed which follows two different distribution forms of anchor nodes.A small WSN in real scene is constructed,in which CC2530 is used as the main control chip and other peripheral circuits.Also,the existing matured Z-Stack protocol stack and ZigBee protocol are incoparated to realized the communications between nodes.Based on the contructed localization prototype system,the proposed static and dynamic node localization algorithm experiments are done the actual indoor empty environment.The experimental results show that the localization rate,position accuracy and positioning efficiency of the proposed algorithms in this paper is the same as the simulation resuts.It can be concluded that the proposed static and dynamaic localization algorithms in this paper can be used in practical applications with satisfied performances.