Scribd
Upload
28 views1 page

AWSN Unit-VI 19

The document describes three schemes for sensor nodes to locally determine whether they are within a radius r of an edge. The schemes are: 1) A statistical scheme where nodes collect neighbors' positions and predicates and compute statistics to count neighbors with true or false predicates. 2) A scheme inspired by image processing using a modified high-pass filter. 3) A classifier-based approach that assumes edges can be approximated by straight lines, and tries to find a line maximizing sensors with the same predicate on either side.

Uploaded by

Akshay bhonde
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
28 views1 page

AWSN Unit-VI 19

The document describes three schemes for sensor nodes to locally determine whether they are within a radius r of an edge. The schemes are: 1) A statistical scheme where nodes collect neighbors' positions and predicates and compute statistics to count neighbors with true or false predicates. 2) A scheme inspired by image processing using a modified high-pass filter. 3) A classifier-based approach that assumes edges can be approximated by straight lines, and tries to find a line maximizing sensors with the same predicate on either side.

Uploaded by

Akshay bhonde
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

 Sensor nodes can be arbitrarily deployed, but each such node knows its location, perhaps using a

localization system
 Three different schemes by which a node can locally decide whether it is within some radius r>0
of an edge or not.
 All these schemes base their decision on locally obtained information. Specifically, a nodes
collects from all other nodes t within some radius R>r of s their respective positions (xt,yt) and the
value et of t’s event predicate. This information is combined with s’s own event predicate. The
radius R is called probing radius and is a measure of the size of the neighborhoods considers and
thus a measure of the communication overhead, which scales roughly as O(R2)for a sensor field of
homogeneous density. The three schemes are as follows.
 A statistical scheme: nodes collects the neighbors’s positions and event predicates and
computes some statistics of these values. As an example, s could count the number NT of
neighbors (including s) for which the event predicate evaluates to true and the number NF
of neighbors with value false.
 A scheme inspired by image processing: a modified high-pass filter with weights
considering the arbitrary node placement is used and the resulting spatial variation in both
x and y is compared against a threshold value.
 A classifier-based approach: it is assumed that the edge/contour is “large” compared to R
and that it can locally be well approximated by a straight line. The classifier running in
nodes tries to find a line such that the number of sensors having the same event predicate
value on either side of the line is maximized. If the resulting line has a point with
distance<r to node s, then s assumes to be on the edge; otherwises is an interior or exterior
node

You might also like