Research On Detection Method Of Selective Forwarding Attack Based On Wireless Sensor Networks

Wireless Sensor Networks(WSNs)are a new type of technology that consist of a large number of micro-sensor nodes for data acquisition,processing and transmission via wireless communication.With the rapid development and maturity in recent years,the technology is not only limited to the military field,but also has great potential in other fields such as civilian use.Therefore,it is particularly important to ensure the safe transmission of data in WSNs.As a common internal attack type in WSNs,selective forwarding attacks often have a greater impact on the network in combination with other internal attacks.Therefore,detecting selective forwarding attacks is critical to the security of WSNs.The specific research done in this paper is as follows:Firstly,the concept and characteristics of WSNs and their development status are introduced,and the selective forwarding attacks commonly found in WSNs are introduced.At the same time,the two common WSNs intrusion detection models are analyzed in detail for the threats caused by selective forwarding attacks to WSNs data leakage.Secondly,aiming at the selective forwarding attack behavior in wireless sensor networks,a selective forwarding attack detection method based on multi-hop acknowledgment and trust evaluation(MHA-TE)is proposed.By sending request packet through source node,the intermediate node replies to the response packet to determine the node that generates the malicious packet loss behavior in the path.Then the Bate distribution is used to determine the trust value of each node in the path.The updated trust value is compared with the corresponding trust value threshold to determine the malicious node.The model solves the problem that the multi-hop confirmation model has low detection rate of multiple malicious nodes and the static adaptability of the trust evaluation model is poor.Finally,considering the randomness and uncertainty of node forwarding rate in WSNs,it is easy to be affected by many factors such as environment.A selective forward attack detection method based on simplified cloud model is proposed.By introducing asimplified cloud model to the trust evaluation model,using the improved K/N voting algorithm obtains the trust value of the target node.The trust value of the target node is compared with the trust threshold to selectively forward the attack node.The simulation results show that the proposed method can effectively detect the selective forwarding attack nodes generated by the network,and has a high detection rate and a low error detection rate.