Research On Filtering Bogus Data Packets In Wireless Sensor Networks

Wireless sensor networks (WSNs) are believed to be the first of the top ten technologies that will change the world in the future. However, limited computing, memory, communication capabilities, and especially the severe energy constraints present major challenges for such a vision to become a reality. Specifically, WSNs can only be deployed after many key technologies to be solved. Security is one of the key technologies.Sensor nodes are subjected to be compromised, and the comprised nodes can be used to fabricate data, if the bogus data is not checked and filtered, the bogus data will be forwarded to the base station. The attack of the fabricating data not only gets the base station make wrong decision, but also deplete the limited energy of the WSNs. Recently, there are some achievements of research for filtering bogus data, but most of them share common character, all the nodes in the routing path try to check the data packets. If the source is far from the base station, many forward nodes in the routing path, it will consume more energy and increase the delay of WSNs. Therefore, this paper proposed an algorithm based on geographic location, By controlling the authentication region of the nodes, the encrypt key and authentication key are generated based on their position. The node focuses on the nodes, which are in its monitor region, while ignores the other nodes. In this way, the energy depletion decreases and fabricated data can be filtered in a few hops. Furthermore this paper analyses the parameters: local upstream distance: Dis and upstream width: B, which affect the performance of the algorithm, getting which criterions should consider selecting proper values. At the same time, the related papers don't include the security maintenance: key updating and isolating the compromised nodes. Making uses of the secure authenticated broadcast protocol:μTESLA, this paper proposed an algorithm, which can update key and isolate compromised nodes safely, based on the broadcast of base station.Finally, our algorithm and LBRS are realized and compared, based on the geographic routing protocol: GPSR. The results show that if the distance between object and base station in a certain range, our algorithm can drop up to 80% of bogus data within 2.5 hops with a steady filtering rate. While LBRS drop up bogus data in 5 hops, and with distance decreasing between object and base station, the filtering rate become lower gradually. When the distance between object and base station is 360m, the filtering rate is 25%.