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 605 | Next

Yingshu Li, My T. Thai, and Weili Wu

"Wireless Sensor Networks and Applications"

If it does, then make it even simpler.
... keep at it till it gets simple. The whole point of a model is to give
a simplified representation of reality. Einstein once said ???Everything
387
Bhaskar Krishnamachari
should be as simple as possible... but no simpler.??™ A model is supposed
to reveal the essence of what is going on: your model should be reduced
to just those pieces that are required to make it work.???
The methodology employed in these studies can be summarized as follows:
1. Identify the unique functionality of the protocol to be modeled. What
exactly does it do ??” does it provide for routing with data aggregation
from a set of sources to a common sink, or is it for gathering information
from the network to resolve a query for a specific named attribute?
2. Identify the primary performance metric of interest in analyzing this protocol.
For battery-constrained sensor networks this often translates to
minimizing the total number of transmissions required to accomplish the
specified networking task. This is because under the assumption that idle
listening and overhearing can be minimized through appropriate MAClevel
scheduling, transmissions of packets (and their corresponding receptions)
are the primary source of energy consumption.
3. Identify the building blocks of the model. It is typical in these models to
assume some simple topology such as uniform random placement with a
fixed radius for connectivity, or a carefully placed grid of sensors, each
communicating with just their four cardinal neighbors.


Pages:
593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617