Research On The Defensive Of Routing Attacks Using Watermark In WSNS

Wireless sensor network (WSN) is deployed in the unattended area with a large number of sensor nodes for data collection and transmission.It is widely used in the military, industrial,medical and household.Sensor network is a data-centric network. The collected data are transmitted by multi-hops to the base station and are easily attacked by the malicious nodes in the routing process.How to avoid malicious attacks in data forwarding process and guarantee the data security are the primary targets of routing security in wireless sensor network.Selective forwarding attacks and Sybil attacks are the most common and easiest form of attacks in data forwarding process.When a malicious node in data forwarding path initiates a selective forwarding attack, the malicious node will refuse to forward the messages or drop packets randomly, resulting that the base station can not receive the integrated messages.In a Sybil attack, a single node presents multiple identities to other nodes in the network which will destroy the normal routing of the network. Selective forwarding attacks and Sybil attack can be united to destroy the network.This article proposes detection methods of these two routing attacks based on the watermark technology:Digital watermark technology is used to detect the selective forwarding attacks in wireless sensor network.In this method, trust value is used in the routing selected algorithm.Use the improved geographic forwarding algorithm to choose a secure data forwarding path. Watermark is embedded into the original data before data transmission and extracted in the base station.Packet lost ratio can be figured out by comparing the extracted watermark and the original watermark.While the packet lost ratio is abnormal,a hop-by-hop detection is used to detect and address the malicious nodes.This paper also proposes a digital watermark method to detect the Sybil attack. Every sensor node in the network is pre-distributed a key by the base station in the network initialization.Every node can calculate their identity certificate by the key. When the source node starts to send the data to the base station, the nodes in the forwarding path embed their identity certificate into the normal data. The identity certificate can be extracted and compared to the original identity certificate stored in the base station.If the extracted identity certificate does not exist in the certificate list, the node is judged as a malicious node that launches the Sybil attack.In this paper, digital watermark method is used to detect routing attacks in the sensor networks.The watermark message which embedded into the normal data can not bring an increment of the computational amount of the sensor nodes and traffic flow in the network.Extensive experiments prove that these methods can find out the malicious nodes which destroy the data transmission.Even the channel error rate is 10%,the detection rate of these two methods can reach 90%.