Database Reference
In-Depth Information
tree-based overlay network as
Semantic Routing Trees
(SRTs). A SRT
is constructed by flooding the sensor network with the
SRT build request
.
This request includes the attribute (ambient temperature), over which
the SRT should be constructed. Each sensor
s
j
, which receives the build
request, has several choices for choosing its parent: (a) if
s
j
has no
children, which is equivalent to saying that no other sensor has chosen
s
j
as its parent, then
s
j
chooses another sensor as its parent and sends
its current value
v
ij
to the chosen parent in a
parent selection message
,
or (b) if
s
j
has children, it sends a parent selection message to its parent
indicating the range of ambient temperature values that its children are
covering. In addition, it locally stores the ambient temperature values
from its children along with their sensor identifiers.
Next, when Query 2.1 is presented to the root node of the SRT, it
forwards the query to its children and prepares for receiving the results.
At the same time, the root node also starts processing the query locally
(refer
Figure 2.4
)
. The same procedure is followed by all the intermediate
sensors in the SRT. A sensor that does not have any children, processes
the query and forwards the value of
v
ij
to its parent. All the collected
sensor values
v
ij
are finally forwarded to the root node, and then to
the user, as a result of the query. This completes the processing of the
sensor data acquisition query (Query 2.1). The SRT, moreover, can also
be used for optimally processing aggregation, threshold, and event based
queries. We shall return to this point later in Section 4.1.
SELECT
s
j
, v
ij
FROM
sensor_values
s
1
v
i1
s
1
v
i1
s
5
v
i5
s
5
v
i5
s
3
v
i3
s
s
3
v
i3
s
4
v
i4
2
s
2
v
i2
s
s
s
5
v
i5
4
s
4
v
i4
s
3
v
i3
s
s
5
3
Figure 2.4.
Toy example of a Semantic Routing Tree (SRT) and Acquisitional Query
Processing (ACQP) over a sensor network with five sensors. Dotted arrows indicate
the direction of query response. A given sensor appends its identifier
s
j
and value
v
ij
to the partial result, which is available from its sub-tree.
2.3.2 Multi-Dimensional Gaussian Distributions.
The
Barbie-Q (BBQ) system [17, 16], on the other hand, employs multi-
variate Gaussian distributions for sensor data acquisition. BBQ main-
tains a multi-dimensional Gaussian probability distribution over all the
