Research On Node Localization Technology Of Wireless Sensor Network For Precision Agriculture

In order to improve crop yield,crops are heavily irrigated,fertilized and other agricultural activities which breaks the balance of various trace elements in the land.These lead to serious resource waste and environmental pollution problems.Precision agriculture combines traditional agriculture with modern technology by obtaining all kinds of environmental information about crops to monitor and manage crops.According to their different growth needs,not only increases the yield,but also reduces the waste of resources.Wireless sensor network(WSN)is a multi-hop self-organizing network composed of a large number of micro-sensor nodes in the monitoring area.WSN has the characteristics of low cost,low power consumption,real-time,strong adaptability.So WSN has wide application prospects in the field of precision agriculture.First,the sensor nodes are randomly scattered in the monitoring area,then the required information is collected and sent to the server.This information must be combined with the coordinates of the sensors to be useful.But it is unrealistic to install GPS(Global Position System)on all sensors to obtain position coordinates,which greatly increases costs.This dissertation mainly studies the problem of positioning static nodes and dynamic nodes in WSN.Based on the principle of localization algorithm,the causes of error are analyzed to improve the localization accuracy of nodes.The main work and innovations are as follows:1.According to the basic principles of the standard DV-Hop algorithm,the conventional DV-Hop optimization algorithms are classified and summarized.These algorithms are introduced in the following three steps: average hop distance,distance estimation and location solution.In each step,several selected algorithms are analyzed and compared in theory.Through a large number of simulation experiments,the better optimization strategy is obtained by comparative analysis,in order to provide guidance for subsequent research.2.In the DV-Hop algorithm,the optimization of multilateral localization is one of theimportant factors affecting the localization accuracy.The least squares method is simple and easy to implement,but its accuracy is low.The standard particle swarm optimization(PSO)algorithm has higher accuracy than the least squares method,but it is prone to premature and has poor global optimization ability.In order to solve these problems,this dissertation proposes a parameter random sampling particle swarm algorithm with enhanced convergence(SC-PSO).Since the optimization effect of PSO is related to the control parameters,the SC-PSO algorithm not only increases the flexibility of the selection,but also enhances the randomness of the particle speed and position update by randomly sampling the control parameters.The sampling range for inertial weight is determined after the acceleration factors have already been sampled in their respective value interval to ensure convergence for every evolution step of algorithm.Besides that,in order to make full use of dimension information of some better particles,the method of updating dimension by dimension is used to optimize the optimal value of population.Finally,the SC-PSO is applied to the DV-Hop algorithm.Experimental results show that the proposed algorithm improves the accuracy of the position solution.3.In order to improve the accuracy of distance estimation between nodes and least squares method,this dissertation analyses the causes of localization error of standard DV-Hop algorithm and presents a DV-Hop localization algorithm based on anchor node selection.First,an intermediate anchor node is selected to calculate the first distance between an unknown node and a target anchor node.Then,the weighted sum of the first distance and the second distance obtained by the DV-Hop algorithm is used as the final distance.Finally,the weighted least squares method is used to calculate the coordinates of unknown nodes.The experimental results show that the proposed algorithm effectively improves the localization accuracy.4.The Monte-Carlo Localization(MCL)algorithm is a common dynamic localization algorithm.The localization principle of MCL is introduced,and the cause of location error is analyzed.In order to improve sampling efficiency and localization error of MCL algorithm,this dissertation proposes a MCL algorithm based on cubic spline.The proposed algorithm shortens the sampling range by predicting the coordinates of node,in order to improve thesampling accuracy of the algorithm,and reduce the running time of the algorithm.