Localization Accuracy Optimization Algorithm For Wireless Sensor Networks Based On Mobile Sinks

With the rapid development of micro circuit integration technology,sensor technology and Internet of Things,the research of Wireless Sensor Network technology is becoming more and more mature.Wireless sensor networks with self-organization are consisted of a large number of nodes,which can collect information,deal with and transmit data and transfer the messages with cooperation to the terminal.Wireless network has easy deployment,dynamic change of the topology and other advantages,which make it plays an important role in the field of industrial production,intelligent agriculture and environmental monitoring.As the basis of the research of Wireless Sensor Networks,localization technology is becoming the focus of the theoretical and applied research.Positioning technology is mainly classify as ranging and range free.Based on the hardware module,the range methods can achieve the better positioning accuracy.However,due to the cost and energy consumption,based on the range free algorithm getting more research significance,has become a research hotspots.This thesis proposes an improved algorithm based on DV-Hop and the improvement of path planning based on mobile anchors.And the details of improvement are as follows:(1)It analyzes the defects of DV-Hop and modifies the single hop distance with the strength of RSSI.At first,it needs to gather some groups of RSSI signal.And according to the laws of the normal distribution,the elements with larger deviation can be removed to enhance the accuracy of RSSI strength.Furthermore,the average hop distance is calculated out by DV-Hop,and the signal intensity of RSSI is modified according to the signal strength of each single hop.(2)Based on the ratio of the hop value and the modified hop value,the localization algorithm can be analysis by classification.When the ratio is lower,the transmission path between the unknown nodes and the path between the anchors are approximately as the same and it can see the product of the average hop size and the modified hop count as the distance.When the degree is lower,it needs to adjust the value of a single hop,by introducing the weight to avoid error with long distance and improve the accuracy of the distance between the anchor and the unknown nodes.(3)In consideration of the disadvantages of the static localization,the improvement of mobile localization is proposed.Based on the cellular model,the improved threefold-coverage localization is used to realize the localization of node coordinates.Two messages with larger strength can be picked out of from the three messages and it can filter out the result with larger error.Then from two messages we can get the two results of the position of the unknown node.And according to the three messages,the final result can be received.(4)The improvement measures are proposed based on the existing mobile path planning.The central and the border anchors are applied to visit the relevant positions of Hexagon-based model with corporation and broadcast the signal at the same time.It can make sure that every unknown node can get three messages at least.In addition,it can get rid of the redundant path and the messages in the same line,so it can improve the accuracy of the localization.