Abstract

Node localization has long been established as a key problem in the sensor networks. Self-mapping in wireless sensor network which enables beacon-based systems to build a node map on-the-fly extends the range of the sensor network's applications. A variety of self-mapping algorithms have been developed for the sensor networks. Some algorithms assume no information and estimate only the relative location of the sensor nodes. In this paper, we assume a very small percentage of the sensor nodes aware of their own locations, so the proposed algorithm estimates other node's absolute location using the distance differences. In particular, time difference of arrival (TDOA) technology is adopted to obtain the distance difference. The obtained time difference accuracy is 10ns which corresponds to a distance difference error of 3m. We evaluate self-mapping's accuracy with a small number of seed nodes. Overall, the accuracy and the coverage are shown to be comparable to those achieved results with other technologies and algorithms. © 2012 IEEE.