In addition to mobility, time
varying fading and shadowing due to changes in the surrounding environment
may also result in a dynamic network topology, even when the nodes are static.
In networks where data updates are to be sent to the sink infrequently, the
nodes may turn o?® their radio circuitry to save battery power, and enter a
sleep state. When one or more nodes enter a power saving mode, the topology
of the network changes. Network topology also changes due to node failures.
Thus sensor networks often have a dynamic topology due to node mobility,
time varying channel, node failures, and radio duty-cycling. If the nodes are
highly mobile, then the node mobility has a stronger impact on the network
topology than the other factors. This has to be taken into account in designing
communication protocols.
4 Common Design Problems in Sensor Networks
Although sensor networks have di?®erent characteristics in terms of their applications
and requirements, there are certain design problems that are common
to most of the sensor networks. These problems are as follows:
4.1 Communication versus Computation
Communicating information over the wireless channel is orders of magnitude
more energy-intensive than computing [2]. In several sensor network applications,
it is possible to perform local signal processing within the network to
compress or aggregate the gathered data. Any form of in-network data aggregation
can reduce the amount of data that is actually sent to the sink, and
this results in considerable energy savings.
Pages:
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52