Research On Three-dimensional Node Localization Technology For Wireless Sensor Network Based On RSSI-AOA And LSSVR

Wireless sensor networks(WSN)have been widely used in military defense,disaster relief,environmental monitoring and urban management.In these applications,the geographic localization information of nodes in the network is needed to know the exact localization of the information source.The node without localization information is a node with no application value.Therefore,the node localization problem is an important research direction in WSN.In practical applications,nodes are not only distributed in a three-dimensional space environment,but nodes may also be mobile.The existing node localization algorithms are mostly oriented to the two-dimensional space range,and are more targeted to static unknown nodes.Therefore,studying the static and mobile node localization of WSN has important theoretical value and application value.Aiming at the above problems,this paper proposes a WSN three-dimensional node localization algorithm based on Received Signal Strength Indication Ranging Algorithm(RSSI),Angle of Arrival(AOA)and Least Squares Support Vector Machine(LSSVR).The main work of this paper includes the following:The paper expounds the research background of this topic,the research status at home and abroad,the purpose and significance of the research,and then briefly expounds the system structure of WSN,the hardware structure of sensor nodes and the characteristics and application of WSN.On this basis,the current WSN node ranging algorithm and positioning algorithm are elaborated and analyzed according to different classification criteria,and the performance evaluation indicators of the positioning algorithm are introduced.Aiming at the problem of static node localization in 3D environment,this paper proposes an LSSVR localization algorithm combining RSSI ranging and AOA ranging.The proposed algorithm introduces AOA angle ranging information into the RSSI ranging information,with multiple anchor nodes as the center of the circle,the wireless range is the circle of the radius,and the coincident part of the circle is the localization area of the unknown node,after the coincidence of the RSSI ranging based on the coincidence of the AOA ranging based overlay,the area of the coincident portion is significantly reduced,and the ineffective area is eliminated.On this basis,the ranging information that combines the distance,azimuth and elevation angle between nodes is modeled as the input of LSSVR to realize the positioning of unknown nodes.Matlab simulation results show that compared with RSSI and LSSVR based on AOA and LSSVR,RSSI and AOA based node localization algorithms are compared,the three-dimensional static node localization algorithm based on RSSI-AOA and LSSVR proposed in this paper can effectively reduce the positioning error and improve the positioning effect.Aiming at the problem of mobile node localization in 3D environment,starting from the multi-hopping of WSN,the motion model of the mobile node is constructed by using the shortest distance hops between the anchor node and the unknown node.At the same time,the modeling mechanism of LSSVR in the state of mobile node is established.Based on the integration of the distance information between the nodes,the azimuth and the elevation angle,the LSSVR mobile node modeling mechanism is used to effectively locate the mobile unknown nodes by selecting the neighbor nodes that satisfy the shortest distance hops.The experimental results are verified by Matlab.The simulation results show that compared with RSSI and LSSVR based on RSSI and LSSVR,the RSSI and AOA based node localization algorithms are compared,the proposed three-dimensional mobile node localization algorithm based on RSSI-AOA and LSSVR can effectively reduce the positioning error and meet the needs of practical applications.