Networking Reference
In-Depth Information
route is selected on a hop-by-hop basis, the intermediate nodes are involved in the
route selection process when the packet is forwarded to any of them.
Whether route selection is done by source, destination, or intermediate nodes,
the deciding node should depend on one or several metrics for the selection
decision. Determination of which route metric to use is dependent on the appli-
cation requirements and needs. The simplest route metric and the most popular one
is the hop count. The path with the least hop count will be chosen to reduce the
number of intermediate nodes involved in the routing process and so reduce the
control overhead and contention level among nodes. Examples of other routing
metrics include energy consumption level, residual energy of the next hop, QoS
metrics (such as end-to-end delay/jitter, interference level, packet loss rate, link
residual capacity, and load balancing), link security level, and memory con-
sumption. Some of these routing metrics require parameters related to the lower
layers like the QoS-based link quality ones. These parameters can either be passed
from the lower layers to the routing layer, or, in some protocols, this interaction
with the lower layers is done in the form of cross-layer protocol design.
In short, how the route is selected is based on the application/network paradigm
for which the protocol will be used. It is how the route will be selected that
controls the performance of the routing protocol and whether it will satisfy the
needs of the application or not. Route Representation and Data Forwarding
After selecting a route, it should be stored to be followed for data transfer. We
consider both route representation and data forwarding as a single component
as they are highly integrated together and, in many protocols, they are done
simultaneously. Route representation and data forwarding can follow one of two
techniques: exact route and route guidance [ 1 ].
(a) Exact Route
In this technique, the sequence of intermediate nodes that a path should follow
to reach a destination is represented explicitly. There are two approaches for
using the exact route representation and forwarding. These approaches are
routing table and source routing.
• Routing Table
In this approach, each node keeps a routing table where it stores the next hop
to reach potential destinations with one entry per destination. In the proactive
protocols, this routing table contains information and next hops to all other
potential destinations in the network. In the reactive protocols that make use
of the routing table approach, they keep information about the destinations
that they interacted with previously or those nodes that they overheard paths
to them. Also, in these routing tables, they may keep information about nodes
from which they received route requests or route replies for further relaying.
When a packet is to be forwarded, the node looks up the routing table and gets
Search MirCeyron ::

Custom Search