Research On WSN 3D Node Security Localization Based On Improved RSSI-LSSVR

Wireless sensor network(WSN)node localization is a kind of technology that makes use of a small amount of anchor nodes that coordinate position information are known to estimate the position of unknown nodes.Although there are a lot of positioning algorithms,many of the WSN positioning technology only considered the complexity of the environment.Without considering network's security,but WSNs are deployed in the open and unattended environment,criminals may attack the positioning system,theft or forgery the data,which makes the security of positioning facing a great chal enge.The traditional positioning algorithm is difficult to meet the actual demand,whether it's positioning accuracy or positioning security.Therefore,it is necessary to study a more secure,more precise and lower power WSN node localization technology.Firstly,this paper explains the theory of wireless sensor network technology,and the research status of localization technology at home and abroad,and then introduces several typical positioning algorithms and performance evaluation indicators.Through the research on the typical localization algorithm and network security,focusing on the received signal strength(RSSI)algorithm,based on the least square support vector regression(LSSVR)algorithm and Sybil security positioning technology.Due to the complex and changeable positioning environment,there is no accurate mathematical model based on RSSI positioning technology,aiming at the shortcomings of traditional RSSI path loss model,the following aspects are improved.First of al,select the improved channel characteristics of RSSI model,in order to enhance the positioning effect of RSSI path loss logarithm model;secondly,to change the traditional RSSI ranging results averaging approach,multiple sampling RSSI value by the median weighted processing,and through the way to set the threshold value of the location results are screened,thus completing the initial ranging.Then,the least square suport vector regression(LSSVR)algorithm is introduced,which can be used to locate the coordinates of the nodes,and the positioning model of the nodes to be positioned and the distance vector between the nodes to be anchored to the anchor nodes can be obtained by LSSVR.Then,the coordinates of the nodes to be located are calculated by the model of the functional relationship between the coordinates of the nodes and the distance vectors.Considering the RSSI ranging from noise impact is relatively large,while the LSSVR has the noise reduction function,which can reduce the computation and complexity,so combined with the improved RSSI algorithm and LSSVR algorithm to complete the positioning of the nodes,so as to further improve the positioning accuracy and obtain more accurate location of the unknown node.Finally,this paper introduces the technology of witch attack security positioning,analyzes the security problems encountered in the improved RSSI-LSSVR node localization process,especially the witch attacks,and improves the traditional way that through calculating the specific coordinates to detect witch nodes,using the ratio of the received signal strength to determine whether there are any witch nodes.In this way,the ability of the system that resist the attack of the witch will be improved.In this paper,the proposed algorithm is simulated by MATLAB,and the influence of parameter setting on the security positioning accuracy is studied.The accuracy,safety and effectiveness of the proposed algorithm and its application are compared and demonstrated by simulation.Therefore,the algorithm and the improved method proposed in this paper have a positive significance for the research of the node localization technology in 3D Wireless Sensor networks.