In in topology of network and these packets

In this routing protocol, every node in a network keeps has
routing information to every other node which may be situated in any other part
of same network. The information about routes is in various routing tables. The
updating of these routing tables occurs periodically whenever a change in
network topology is sensed. The node that detects change, shares this
information with other nodes in the network. Protocols in this class differs
from each other based on way the updating of routing table take place and also
on the basis of the kind of information that is stored in these routing tables.
The worst-case scenarios for these protocols are represented by their
performance metrics. Following are some proactive routing algorithms: –

Destination-sequenced distance vector (DSDV) is a
proactive table-driven routing protocol that ensures a loop free routing. DSDV
selects a single path from source node to destination node by using shortest
path routing algorithm. DSDV works on updated version of classical Bellman-ford
routing technique. In order to reduce the amount of overhead transmitted
through the network, two types of update packets are used 13P1 .
One packet is used for carrying routing information and other packet carries
the information that is generated due to change in topology of network and
these packets are more exchanged rapidly between nodes within a network. The
delay in this protocol is minimum as every node already knows the path to every
other node. As exchange of messages is required periodically in DSDV, it uses a
significant amount bandwidth available and introduces a large amount of network
overhead. This is the reason why DSDV is not used in scaling large networks
like MANET.  

We Will Write a Custom Essay Specifically
For You For Only $13.90/page!

order now


Wireless Routing Protocol (WRP) also guarantees loops
freedom and it avoids temporary routing loops by using the predecessor
information 14P2 .
In WRP each node in network has four routing tables. Due to every node
possessing four routing tables, when size of the network increases, noticeable
amount of memory overhead is introduced at every node. WRP uses a “HELLO”
message to check the connectivity between two neighbouring nodes when no
transmission has occurred for some time. This practice eats a lot of available
bandwidth and require every node to be in active mode all the time and wastes a
lot of power as nodes cannot enter sleep mode even when no transmission is
going on.


State Routing (GSR)
protocol works on the principle of Link state routing algorithm 15P3 . Global State Routing protocol is
an improved version of link state routing in the way that GSR restricts the
exchange of update message only to intermediate nodes. In GSR, each
participating node holds a link state vector which is updated periodically with
the information obtained from directly connected neighbouring nodes. These
nodes exchange these tables only with their neighbouring nodes. This technique
helps in reducing a large amount of control messages that flows through the
network. However, as the network gets bigger, the size of update messages that
are exchanged between intermediate nodes will also increase which will in turn
waste a large amount of available bandwidth.


State Routing (FSR)
protocol is an improved version of Global State Routing protocol. FSR improves GRS
by reducing the update message size. This is done because FSR allows frequent
routing updates for nodes that are nearby and does not update routing
information of the nodes that are far 16P4 . This enables FSR to scale large
networks, like MANET, better then the protocols previously defined. However,
the ability of this routing protocol to scale large network is compensated by
reduction in accuracy. This happens because as the nodes or mobile stations
within a network start changing their positions, it becomes difficult to find
the routes for destinations that are far. The issue of decreasing accuracy can
be resolved if the frequency at which updates are sent to remote destinations
is made dependent on the rate at which mobile nodes change their locations
within a network.


Tree Adaptive Routing (STAR) protocol is also an improvement of Link state routing algorithm. All
participating routers maintain routing trees. A routing tree is a set of paths
that can be taken to reach a particular destination from that node 17P5 . Source Tree Adaptive Routing
protocol supports Least Overhead Routing Approach (LORA) for exchanging
information between nodes that decreases the amount of routing overhead. Unlike
link state routing algorithm, where routing information was exchanged
periodically, in STAR the information or routing table updates occurs only when
some event is triggered. This property of STAR allows it to be used for routing
purposes in large networks as it consume less amount of available bandwidth.
This also reduces the latency by using predefined paths.   However, as each node maintains a routing
tree, this could introduce overheads for route processing or memory for a large
and dynamically changing network.


Routing Effect Algorithm for Mobility (DREAM) routing protocol follows a different way for
routing information between participating nodes as compared to all the routing
protocols described previously. This protocol uses a Global Positioning System
(GPS) that enables each node to know geographical coordinates of its location 18. P6 Instead of exchanging information
about whole path between nodes, DREAM routing protocol allows each node to exchange
its geographical coordinates with every other node in the network. The
information regarding coordinates is stored in tables called location table.
This practice saves a large amount of available bandwidth from getting wasted
which in turns makes this routing protocol more suitable for large networks.
The rate at which update messages are send is made dependent on mobility of
nodes which results in reduced overhead due to routing. The nodes that do not
change their locations within a network, do not send any update messages to
other nodes.


Support in Mobile Wireless Networks (MMWN) routing protocol uses the concept of
hierarchical clustering for maintaining the network 19P7 . Each cluster consists of two
kinds of nodes: switches and endpoints. Location management in MMWN routing
protocol is carried out by Location Manager (LM), which is in every cluster.
The discovery and updating of location is carried out by Location Manager. This
reduces the routing overhead as nodes in a network do not participate in
exchanging information about routes with each other. Also, this results in
reduction of usage of bandwidth. The other aspect of this routing protocol is
that the discovery and updating of locations of different nodes a difficult
task because to perform this, a message must go through whole tree of the
location manager. There are many issues in this routing technique that make
implementing this protocol a difficult job.


Head Gateway Switch Routing (CGSR) is a hierarchical routing protocol in which nodes within a network are
divided in to clusters. Unlike MMWN routing protocol, CGSR routing protocol
does not require to maintain hierarchical tree. Instead cluster-head is used to
maintain a group of nodes (cluster). In this technique, a node from the cluster
is randomly chosen to be a master node and this chosen node manages all other
nodes in cluster. All the communication between two different clusters
occur through cluster-head and no other node participate in this. The transmission
medium is also controlled by this node. Within a cluster, nodes only maintain
path to their cluster-head, which results in reduction of overhead due to
routing. A noticeable amount of overhead is generated in this routing protocols
in order to manage clusters.


Link State Routing (OLSR) routing protocol work on the principle of link state routing algorithm
and follow point to point routing technique. In this, the network topology
information is maintained by every node in the network through sharing of
routing tables periodically with each other. OLSR routing algorithm implements
MultiPoint Replaying (MPR) which reduces the message size and number of nodes
that broadcasts themselves every time a route is updated.  MPR works as follows: –  When a change in network topology is
detected, a node in a network retransmit its data to the node that are directly
connected to it. This group of nodes is referred to as multipoint relays of
that particular node. Node that are not a part of this group, cannot retransmit
their data, it can only reprocess them. Each node broadcasts a list that
contains all its directly connected nodes using a HELLO message. Every node
present in the list in HELLO message select its directly connected neighbour
which in turn covers all the nodes that are two hops away from initial node.
Every node in the network has a most favourable path to a particular
destination which are determined by network topology information which is
stored in routing tables. The matrices used for these paths are number of hops
taken to reach a destination node from a source node. So, it this algorithm the
paths are defined as soon as communication takes place.


Broadcast Reverse Path Forwarding (TBRPF) routing protocol is also works on the principle
of link-state routing protocol. This protocol is based on the Reverse Path Forwarding
(RPF) technique. Reverse-path forwarding distributes the packet containing information
through a spanning tree from destination side to source side of that message. Spanning
tree contains the least-cost path (in this case minimum hops) to the source from
nodes. Every node defines a route to each destination by using a revised version
of classical Dijkstra’s algorithm. Each node does not share whole information of
their source tree with their directly connected nodes.