Pollution Data Detection Methods Based On Xor Network Coding In Wireless Sensor Networks

Network coding is a new technology of information theory. It can effectively solve the energy saving of wireless sensor networks and improve network performance in many aspects. However, network coding is vulnerable to pollution attacks because of coding. The application of network coding is greatly restricted by the safety problems. Since the energy and resources in wireless sensor networks are limited, in order to improve the network lifetime and against pollution attacks, we study the pollution data detection based on xor network coding in WSNs. The contributions of this paper as follows.(1) An efficient authentication planning for xor network coding is proposed in this paper, which is based on the network lifetime. The presented planning determines node’s mode according to the residual energy of the node. It aims at balancing the energy of nodes and improving the network lifetime. Experimental results show that the network lifetime of overall minimum energy consumption policy is slightly higher in the case of smaller probability of attack, but the presented strategy can improve the network lifetime about11%to48%with the attack probability increasing.(2)For the data pollution in wireless sensor networks, we present a pollution detection scheme in xor network coding systems, which based on probabilistic key pre-distribution and message authentication code (MAC). Each block is attached with a number of MACs and a node can use these MACs to verify the integrity of block with shared key. Due to exclusive or homogeneity of MAC, the node’s communication overhead is unrelated to the number of input. Compared to previous scheme, the presented scheme can reduce the communication overhead.(3) Combining with the elliptic curve mechanism and homomorphism hash function, we design a xor network coding signature scheme which can identify malicious node. Source compute the redundancy of original information by private key, and the public key is equal to the hash value of new information. The intermediate nodes verify the integrity of the received packet. Unlike the traditional network coding signature scheme, the presented scheme use addition and multiplication on the finite field GF(2q). It is suitable for the processing of the sensor nodes and can reduce the computing burden of sensor. The presented scheme can filter the pollution as soon as possible and identify the malicious node. It can provide the path selection for data transmission and improve the security and network performance.