Research On Node Location Technology In WSN Based On DV-Hop

As a core technology of the sensing layer of the Internet of things(Io T),wireless sensor networks(WSN)has attracted more and more attention with the improvement of researchers’ research heat in the field of the Io T.WSN is widely employed in a variety of industries,including precision agriculture,environmental management,and health care.Most application situations necessitate the position of sensors or devices.Node positioning technology can obtain the location of sensors or devices,so positioning technology plays a very important role in WSNs.The DV-Hop algorithm is a widely used localization approach.It has the advantages of independent ranging technology,low positioning cost and low complexity.Therefore,there is significant research value in the DV-Hop algorithm.In this thesis,the positioning principle of the DV-Hop algorithm is discussed,its positioning error sources are theoretically investigated,and several improvement techniques to decrease positioning error are presented.The main work contents are as follows:To reduce the localization errors caused by the hop count and hop distance calculation methods,the DV-Hop localization algorithm based on hop distance weighting is proposed.In this method,the individual hop values are first subdivided using a double communication radius to obtain more accurate hop values;then the distance between nodes is calculated separately using different average hop distances in the range of different node hop values,and the ratio of hop distance and communication radius is used as a new weight for controlling the influence of different anchor nodes on the hop distance;finally,the node position is calculated using least squares method.Through simulation experiments,it is shown that the improved strategy can improve the positioning accuracy of the algorithm.To reduce the localization error caused by solving the unknown node position by the least square method,a DV hop algorithm based on Golden sinusoidal particle swarm optimization is proposed.The method uses the golden sine particle swarm algorithm to estimate the node positions.The golden sine particle swarm algorithm is an improvement of both the inertia weight and the position update method.Firstly,an exponential function is used to achieve a non-linear decreasing inertia weight,and a random number is added to adjust it dynamically.Finally,the particle position is updated according to the principle of the golden sine optimization algorithm.The results of the test function prove that the Golden Sine particle swarm algorithm has good optimization finding capability.Under the same simulation conditions,the proposed improved DV-Hop algorithm effectively reduces the localization error.The DV-Hop algorithm based on mobile anchor nodes is proposed to reduce the cost of localization and the localization errors caused by the uneven distribution of nodes.The method uses a mobile anchor node to localize unknown nodes in the network according to a planned path.The main work is to select the locations of the virtual anchor nodes using the improved Antlion Optimizer and to obtain the shortest movement path between these virtual anchor nodes according to the Traveling salesman problem.The improved Antlion Optimizer proposes to update the ant positions using the black hole strategy and backward learning,which can improve the convergence speed and population diversity of the algorithm.The test function results show that the improved Antlion Optimizer has good optimization finding capability.Under the same test conditions,the proposed path planning based on moving anchor nodes has better performance in terms of the number of virtual anchor nodes,path length,and localization accuracy compared with other static path planning.