MLDS: A flexible location directory service for tiered sensor networks

  • Authors:

  • Sangeeta Bhattacharya;Chien-Liang Fok;Chenyang Lu;Gruia-Catalin Roman

  • Affiliations:

  • Department of Computer Science and Engineering, Washington University in St. Louis, One Brookings Drive, Box 1045, St. Louis, MO 63130, USA;Department of Computer Science and Engineering, Washington University in St. Louis, One Brookings Drive, Box 1045, St. Louis, MO 63130, USA;Department of Computer Science and Engineering, Washington University in St. Louis, One Brookings Drive, Box 1045, St. Louis, MO 63130, USA;Department of Computer Science and Engineering, Washington University in St. Louis, One Brookings Drive, Box 1045, St. Louis, MO 63130, USA

  • Venue:
  • Computer Communications
  • Year:
  • 2008

Quantified Score

Hi-index 0.24

Visualization

Abstract

Many location based services, such as a those used in healthcare facilities to track medical personnel and equipment, need to keep track of mobile entities across wide areas. These services can be realized in a cost-effective and efficient manner using tiered sensor networks comprised of multiple wireless sensor networks connected by an IP network. To simplify the realization of such services, we present MLDS, a Multi-resolution location directory service for tiered sensor networks. MLDS provides a rich set of spatial query services ranging from simple queries about entity location, to complex nearest-neighbor queries. Furthermore, MLDS supports multiple query granularities which allow an application to achieve the desired tradeoff between query accuracy and communication cost. We implemented MLDS on Agimone, a unified middleware for sensor and IP networks. We then deployed and evaluated the service on a tiered sensor network testbed consisting of tmotes and PCs. Our experimental results show that, when compared to a centralized approach, MLDS achieves significant savings in communication cost while still providing a high degree of accuracy, both within a single sensor network and across multiple sensor networks.