Research On Data Authentication Model Based On RS Erasure Code In Wireless Sensor Networks

Wireless sensor network attracts more and more attention by its unique advantage. Now it is widely applied to military use and civilian use. But the sensor node is limited in its energy, computation, and communication capabilities. It is vulnerable to enemy attack. So, how to guarantee the security of data transmission is a hot topic. In the open WSN environment, data authenticity is one of the important attributes in data security. It is facing grate challenges, including packets discarded and doctored maliciously by adversary. How to verify the authenticity of the received packets in the enemy environment becomes an important research topic in WSN data authentication.For the existing problems of the packet losing and tampering in WSN, we propose an authentication model based on Reed-Solomon (RS) erasure code, signature amortize and elliptic curve signature technology. We call it Data Authentication Model based on RS Erasure Code (DAM-RS). First we define the model of WSN which is controlled by an adversary (WSN-CA). This model allows data losing and tampering. Through the definition of packet survival rate a limiting the ability of adversary, so that the network model is closer to the actual network environment. Based on WSN-CA, the formalized definition of the DAM-RS is given. We also give the entity communication, the authentication algorithm and the verification algorithm of the DAM-RS. Analyzing the correctness of the model, the result shows that when the packet survival rate more than a, all of the valid packets will pass the verification, and it can recover the data packets which are deleted by the adversary. Through the simulation of the enemy attack to analyze the safety of DAM-RS, it shows that DAM-RS can resist deleting and tampering attack.Finally, we compare the DAM-RS with the current typical WSN data authentication. The result shows that DAM-RS not only provides the ability to prevent data from being deleted and tampered, but also has lower computation cost, communication cost and storage cost. So it is suitable for the application of WSN. At the same time, we simulate each model in TOSSIM.