Research On Communication Security In Cluster-based WSN

With the development of the Internet of Things(Io Ts)technology and the demand of industry,Wireless Sensor Network(WSN)have been widely concerned.Cluster-based WSN are often employed in multi-hop transmission of large-scale WSN to alleviate data implosion,reduce transmission delay and improve energy efficiency.Clustering schemes divide sensor nodes into clusters according to certain principles.In each cluster,one head node is used for intercluster communication,and the other nodes in the cluster are used to collect information and combine the information to the head node.WSN brings us many advantages and convenience,but the hardware characteristics of wireless sensor nodes make the communication between sensors not only have the openness and broadcast nature of wireless transmission,but also be limited by the low power and constrained energy of sensor equipment.The former makes the confidential information in communication easy to be disclosed,while the latter makes it difficult to deploy the security scheme in networks.These together lead to security problems in WSN.Therefore,to solve the security problems in Cluster-based WSN,it is particularly important to propose reasonable and effective security schemes in Clustering scenario.At the same time,advanced network analysis models are so useful to conduct performance analysis(including interference and security analysis)for Clustered WSN that we can better define the network and design security solutions and amplify the advantages of Clustered WSN.This paper firstly proposes a cooperative jamming scheme based on physical-Layer Security(PLS)for clustered WSN.The scheme takes advantage of the natural topology of Clustered WSN and employs normal nodes in clusters to jointly send artificial noises to interfere with illegal eavesdroppers,thus improving the security of inter-cluster communication.These artificial noises are intentionally extended through orthogonal vectors,so that the legitimate receiver can effectively eliminate the influence of the noises by taking advantage of the their orthogonality.This scheme allows the legitimate receiver to keep strong security even without location or channel advantages over the eavesdropper.We verify the security performance of the scheme from the point of secrecy rate through extensive simulations.Furthermore,we use stochastic geometric model to conduct interference and security analysis for clustered WSN.The model assumes that nodes for all roles are randomly distributed in their particular regions.In this case,we propose a general analysis framework based on stochastic geometric model to analyze both inter-cluster interference and physical layer security performance from two aspects of link capacity(LC)and security capacity(SC),respectively.Two scenarios are considered in this paper,i.e.,interference scenario and eavesdropping scenario.The rationality of the proposed framework is verified by plenty of simulations,and the influence of different network parameters on the performance metrics is demonstrated.The analysis framework provides insights for parameter setting of Clustered WSN from the perspective of interference and security.Finally,we take the cooperative jamming scheme into primeval clustered WSN and analyze the network as a whole using the proposed general analysis framework.The analysis results not only verify that the proposed cooperative jamming scheme greatly improves network security performance,but also show the influence of different scheme parameters on security performance,which conversely provides parameter suggestions for our cooperative jamming scheme.