??? Distributed region determination: Distributed algorithms have always been
flexible, and thus can be used to detect and track large-scale phenomena
autonomously and dynamically.
Fig. 1. A sensor network for tracking chemical plumes.
The rest of this chapter is organized as follows. Section 2 presents several
classical localized edge detection schemes. Section 3 provides an overview of
centralized edge determination and a simple algorithm. Section 4 includes
some key algorithms for hierarchical edge estimation. Section 5 provides an
overview for distributed region detection algorithms. And finally, Section 6
summarizes this chapter with pointers to some future research directions on
boundary detection for sensor networks.
150
Chapter 6 Boundary Detection for Sensor Networks
2 Localized Edge Detection
Localized edge detection is a technique by which each sensor node determines
either locally or by collecting information from other nodes within its neighborhood
whether or not it is an edge sensor. Ideally, a sensor with positive
value of event predicate and situated right on the boundary is said to be an
edge sensor. However, this is only a description rather than a definition. To
define an edge sensor, the notion of tolerance radius is introduced. As shown
in Figure 2, if a sensor node resides in the event region and there is an edge
passing through its tolerance radius, then, it is regarded as an edge sensor.
Pages:
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265