SEARCH
0-9 A B C D E F G H I J K L M N O P Q R S T U V W X Y Z
Prev | Current Page 320 | Next

Yingshu Li, My T. Thai, and Weili Wu

"Wireless Sensor Networks and Applications"

In addition, the control
and energy overhead of the switching itself has been neglected, as is commonly
done. However, some work has investigated the e?®ect of the switching overhead
as well [19] [11].
The above assumptions are typically reasonable first-cut approximations,
as can be seen in Figure 1. In this case, the average energy consumption,
normalized versus a scenario where a node never transitions to the sleep mode,
is given by:
Enorm ??
Tperiod
Tactive
. (7)
By combining equations (6) and (7), the normalized energy consumption
in the quasi-dormant state can be expressed as a function of the normalized
worst case wakeup delay:
Enorm ??
1
Dnorm + 1
. (8)
This equation expresses the fundamental tradeo?® between energy and delay
that is present in the majority of proposed wakeup solutions. Figure 4
illustrates the nature of this tradeo?®. By allowing a small delay, large energy
savings can be obtained initially. The exact behavior and shape of the tradeo?®
curve depend on the specifics of protocol [13]. Instead of analyzing various protocols
in detail, this fundamental tradeo?® provides a solid and easy method
for a rough comparison between them. Essentially, the tradeo?® behavior of
equation (8) underlies most of them, and the main di?®erence between them
is the minimum value of Tactive, which maps the normalized wakeup delay
to the absolute one (which is the actual constraint). Solutions with a smaller
Tactive are generally preferable, although other factors also come into play, as
will be detailed further in the later sections of this chapter.


Pages:
308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332