Research On 3D Localization Algorithm In Complex Environment Based On Wireless Sensor Networks

As an important branch of computer networks,wireless sensor networks,which integrate perception,acquisition and monitoring technology,have been widely concerned by the researchers since their birth.With the features of WSNs,they have been widely applied to the field of environment monitoring.In actual environmental monitoring,the collection and analysis of environmental parameters by wireless sensors must be based on knowing the location of the node itself.If the location of the node cannot be obtained,all the information is meaningless.Therefore,designing low-cost and high-reliability localization algorithms,especially 3D localization algorithms for complex outdoor environment,becomes a hot research focus.On the basis of studying a great deal of previous research work,this thesis started with the 3D DV-Hop algorithm,a commonly used range-free localization algorithm.The source of the positioning error and the influencing factors of positioning accuracy of the algorithm were analyzed in detail through theoretical analysis and simulation experiments.A 3D DV-Hop-RE optimization localization algorithm was proposed for these problems.The details are as follows:(1)Aiming at the error causes in the distance calculation of the 3D DV-Hop localization algorithm,corresponding improvement methods were proposed in this thesis according to the error causes obtained through analysis.First,the method of subdividing the minimum hop count was proposed.The original one hop was divided into smaller hop counts according to the subdivision coefficient,so that the hop count between neighboring nodes was more consistent with the actual distance.Second,referring to the advantage of RSSI localization algorithm,the signal strength received by the node was included in the calculation.The hop count of neighboring nodes was adjusted according to the number of the received signal strength.Besides,the minimum mean square error criterion was introduced to adjust the calculation formula of the average hop distance of beacon nodes.When calculating the distance between unknown nodes and beacon nodes,the calculation of the average hop distance was improved with hop count as the weight to make the average hop distance of the unknown node more representative of the network.A concept of node threshold was proposed to reduce the error.,which was used to limit the number of beacon nodes participating in the calculation.(2)Aiming at the problem that the 3D DV-Hop localization algorithm cannot localize due to the irreversibility of the matrix in the position calculation formula,the particle swarm algorithm was proposed to replace the original maximum likelihood localization method.To improve the accuracy and efficiency of the PSO algorithm,the fitness formula was redesigned to avoid the error caused by the bending of the node path.The fixed learning factor was changed to be dynamic to adapt to the requirements of different phase for particle iteration speed and direction.At last,through simulation calculations and field tests,the effects of the optimization algorithm were verified in terms of the nodes number,the maximum communication radius and the proportion of beacon nodes.The result showed that the 3D DV-Hop-RE optimization algorithm proposed in this thesis has obvious advantages in localization accuracy compared with the 3D DV-Hop localization algorithm.