Research On Routing And Positioning Technology In Wireless Sensor Network

Wireless sensor network(WSN)has been deeply widely used in the national economy and daily life and is responsible for many functions such as data acquisition,status measurement,and equipment positioning.Wireless sensor networks are composed of a large number of nodes connected by wireless self-organizing.According to the predefined routing protocol,each node independently acquires data to transmit the data to the central server.In view of the characteristics of wireless sensing networks in different applications,a variety of routing protocols have been developed,with different emphasis on node energy consumption,data latency,congestion control,and other aspects of performance.The geographic routing protocol is one of the more important ones,which mainly uses the geographic or location information of the node in routing decision-making,which can effectively control node energy consumption and quickly cope with network topology changes.The basis of the geographic routing protocol is the node location information.The positioning algorithm based on the wireless signal fingerprint has low cost,higher positioning accuracy,and is suitable for node positioning in the wireless sensor network.This paper focuses on congestion control,network energy efficiency,and node positioning in wireless sensor network and proposes corresponding solutions.The main work of this dissertation is as follows:The sudden traffic generated by abnormal events in the monitoring area of the wireless sensor network will cause network congestion,resulting in increased data packet latency and increased energy consumption of the node.The congestion seriously affects the quality of service of the network.For the congestion problem of energy-constrained WSN,a congestion control routing algorithm based on gravitational competition is proposed.The transmission gravity and the energy gravity are defined according to the distance and position relationship between neighbor nodes and the node cache state.Transmission gravity describes the node's congestion state from three aspects:the data packet arrival rate,the cache queue length,and the data packet processing capacity.The energy gravity estimates the remaining energy of the node by the distance and considers the energy distribution of downstream nodes.According to the principle of work in physics,the gravitational competitive intensity is introduced to express displacement.The node gravity and the gravitational competitive intensity are multiplied as work.The node with the largest work is selected as the next-hop node.In addition,for the congestion problem of the WSN with supplementary energy function,a traffic-aware routing algorithm based on virtual force is proposed.The algorithm defines the virtual internal force based on the distance between the source node,the neighbor node,and the destination node to build the maximum single-hop distance of the packet.It defines the virtual external force according to the cache queue length to perceive the forward regional traffic and construct a route with idle or low-load nodes.The virtual external force and the virtual internal force merge into a virtual integrity force,driving packets to bypass the congested region and transfer them to the destination node as soon as possible.For energy-constrained sensor nodes,improving energy efficiency can extend the WSN lifetime and enhance network availability.The geographic routing algorithm can select the optimal transmission power consumption based on the node distance.Thus.a multi-parameter fusion energy-efficient routing algorithm is proposed.The algorithm defines five routing evaluation parameters:the node remaining energy,effective forwarding rate,single-hop transmission ratio,cache queue index,and energy equilibrium.Each parameter reflects the state of the network or nodes from a different perspective.The credibility coefficient is introduced to evaluate the credibility of each parameter in the routing decision.The parameter contribution degree reflects the important difference between each parameter.The fuzzy contribution degree is introduced for the distance measurement error caused by random interference in the network.The parameter contribution degree and the fuzzy contribution degree are combined into the fusion contribution degree.The node with the maximum fusion contribution degree is selected as the next-hop node.The geographical routing decision requires accurate node location information.In addition,it is also necessary to accurately locate a node in wireless sensor network applications,such as logistics and medical and other fields.The range-free fingerprint positioning algorithm has lower node capability requirements,shorter positioning time,higher accuracy,and lower deployment cost among many positioning algorithms.In recent years,it has received great attention in the academic and industrial.The fingerprint positioning includes two stages.The offline measurement stage establishes a database of the relationship between the wireless signal fingerprint and location coordinates of the positioning area;the online query stage estimates the coordinates of unknown nodes according to a specific algorithm.In this dissertation,two fingerprint positioning algorithms are proposed,respectively,for single-region and multi-region.For the node positioning in a single-region,a fingerprint positioning algorithm based on Gaussian distribution is proposed.The fuzzy method fuses unknown node fingerprints and database fingerprints to form a fuzzy decision matrix.The algorithm uses the number of anchor nodes as the dimension to establish a multi-dimensional Gaussian distribution model.uses the Mahalanobis distance to estimate the deviation of the unknown node with reference nodes,and selects the K reference nodes with the smallest deviation to estimate the coordinates of the unknown node.A fingerprint positioning algorithm based on fuzzy decision-making is proposed in accordance with the knowledge decision theory for positioning in a multi-regional complex environment.The positioning process is divided into three phases:in the knowledge accumulation phase,fuzzy fusion is used to form a positioning decision matrix;in the knowledge fusion phase,the Lagrangian optimization algorithm is used to obtain the membership weights of anchor nodes in different regions;in the knowledge expansion phase,the algorithm performs positioning decision,calculates the matching degree of the reference point,and uses the coordinates of the reference point with the highest matching degree as the estimated value of the position of the unknown node.This dissertation focuses on the congestion control,energy efficiency optimization,and node positioning of the wireless sensor network.Based on the physical distance between nodes,routing algorithms and fingerprint positioning algorithms are proposed.The effectiveness of the proposed algorithms are verified through theoretical analysis and software simulation.