Authors: Ki Young Lee; Jeong Jin Kang; Jeong Joon Kim; Gyoo Seok Choi; Young Dae Lee; Seong Yun Cho; Sun Jin Oh
Addresses: Department of Medical IT and Marketing, Eulji University, Seongnam 461-713, Korea ' Department of Information and Communication, Dong Seoul University, Seongnam 461-714, Korea ' Division of Computer Science and Engineering, Konkuk University, Seoul 143-701, Korea ' Department of Computer Science, Chungwoon University, Incheon 402-803, Korea ' International Promotion Agency of Culture Technology, Seoul 138-169, Korea ' Department of Digital Systems Engineering, Anyang University, Anyang-si 430-714, Korea ' Department of Computer and Information Science, Semyung University, Jecheon 390-711, Korea
Abstract: Most of previous studies attempted to solve problems resulting from the frequent update of moving sensor node data by extending R-tree, the representative space index structure. Some recent studies consider the distribution of moving sensor nodes using histogram as an additional structure. Nevertheless, it is not easy to process non-uniform distribution because in such related works uniform distribution is processed simply based on histogram of fixed split. In this paper, we propose RGP-tree (R-tree with dynamic non-uniform grid and sub R-tree), which integrates the dynamic non-uniform grid method and the sub R-tree method. Because RGP-tree can split space dynamically and non-uniformly even if moving sensor nodes are distributed densely in a specific area, it is possible to process moving sensor node data effectively.
Keywords: sensor networks; moving sensor nodes; dynamic non-uniform grid; frequent updates; R-tree; node mobility; indexing methods; node retrieval.
International Journal of Sensor Networks, 2014 Vol.15 No.4, pp.238 - 245
Available online: 25 Aug 2014 *Full-text access for editors Access for subscribers Purchase this article Comment on this article