Research On Attack-resistant Node Localization For Wireless Sensor Networks

Advances in micro-electro-mechanical-system and digital electronics have triggered a large quantity of interest in wireless sensor networks(WSNs), which are composed of a large number of simple a low-cost and low-power sensor devices equipped with wireless communication interface in the air. WSNs has been proposed for various potential applications including target tracking,environment monitoring,building and structures monitoring,military surveillance,anti-terrorism,medical treatment,and even traffic surveillance so on and so forth. In those applications, data sensing is more often than not meaningless without relating to its physical location. Furthermore, some middle ware services should pay more close attention, a very good case in point is that the location aided routing needs accurate location information.Then, it is too much emphasized to the localization of sensor nodes, which is the foundation of the other applications and to determine self-position using a certain localization scheme, according to the anchor nodes. At present, almost of all location systems or algorithms more and more devote themselves to acute localization, because of never premeditating the possible security attacks and security requirements but regarding as the truly and friendly environment. During localization of WSNs the captured nods should be able to send error message of localization, this time the location systems are not acute and even do not work, so the fact is how to reality securely and effectively localization sensors coordinates is one of the most important security problems. This thesis that we focus on the problem of self-localization security in WSNs and makes thorough discussion and research of hot problems from global reach in nearly years, and the most important fact is that we give one of attack-resistant node localization algorithm for wireless sensor networks.The content follows that one maximum-likelihood-estimation based attack-resistant localization algorithm could be proposed. Simulations show that the original localization algorithm is unavailable in hostile environments for most existing maximum likelihood estimation algorithms, which the attack-resistant localization algorithm should be able to restore the original maximum likelihood estimation localization algorithm, within RSSI maximum range error of 0.5 communication radius distance.