Many papers that describe protocols for wireless sensor networks mention that the network can be represented as a graph such as unit disk graph. nodes are vertices of the graph and links between nodes are edges of the graph. I need to know if all nodes in the network know this graph in the beginning of the connection? what is confusing for me , from my knowledge i know that node know information about its direct neighbors only. how can i assume in my simulation that network with thousand of sensors , all of them have the graph that represent the whole network???
so if what i think is correct, i.e. the graph is too big and need many communication to be constructed in the beginning of the connection that mean it can't be known to all nodes in the network, what is the benefit that papers always mention that the network represented as graph????
You are missing some points here. Representing the network as a graph eases the application of things like neural networks(which requires a weight, present on graph), or routing(using ping to choose the minor path). On real world, no element need to have access to whole network graph, sink nodes, in maximum, might have a routing table. Graphs are an easy to understand way of the conceptual structure of network.
Hi there...
When you study network as a graph, you are modeling the network to obtain different information. For example, the distribution, cluster of, node degree, etc...These information may be utilized to understand different aspects of the network. so do not miss the point of the researchers when they describe their network as a graph with nodes and edges.
As already pointed out by other researchers Graph Theory is an eefficint way to describe physical entities having interconnection between them. The Graph Theory is already a highly evolved mathemarical tool. So when you visualize very huge network in thr firm of a graph you may easily come to conclusion about many optimization needed in such a network. Without this the mathematical abstraction of the whole network becomes tough. The graph theory implementation helpsalso helps in modeling different phenomenons of WSN. Hope it j
- Badges
- Science topic