Experiences building a distributed sensor network

  • Authors:
  • Victor Lesser

  • Affiliations:
  • Department of Computer Science, University of Massachusetts, Amherst, Amherst, MA

  • Venue:
  • AI'03 Proceedings of the 16th Canadian society for computational studies of intelligence conference on Advances in artificial intelligence
  • Year:
  • 2003

Quantified Score

Hi-index 0.00

Visualization

Abstract

A central challenge in building advanced sensor networks will be the development of distributed and robust control for such networks that scales to thousands of intelligent sensors [8]. Appropriately structuring where and when control and interpretation activities are done is key to the effective operation of the network. This structuring must be adaptive to changing network conditions such as new sensors being added, existing sensors malfunctioning, and communication and processor resource modifications. Together with this adaptive re-structuring of long-term roles and responsibilities, there is also a need for short-term adaptivity related to the dynamic allocation of sensors. This involves allocating the appropriate configuration of sensing/processing resources for effectively sensing the phenomena but also the resolution of conflicting resource assignments that may occur when there are multiple phenomena occurring in the environment that need to be tracked concurrently. More generally, this structuring can be thought of as organizational control. Organizational control is a multilevel control approach in which organizational goals, roles, and responsibilities are dynamically developed, distributed, and maintained to serve as guidelines for making detailed operational control decisions by the individual agents. The parameters guiding the creation and adaptation of the organization can have a dramatic impact on the performance of the sensor network. We have recently completed work on a small-scale sensor network (approximately 36 low-cost, adjustable radar nodes) for multivehicle tracking [5,7], that exemplifies in a simplified form many of the issues discussed above (see Fig. 1). This lecture will discuss how we approached the design of the sensor network and what technologies we needed to develop.