**- The routing algorithm that considers the flow in the network is known as flow based routing.**

- It takes into consideration
the amount of traffic flowing in the network before making a decision regarding
the outgoing link over which the packet has to be sent.

- The ability to
characterize the traffic flow’s nature with respect to time is the key to the
successful implementation of the flow based routing.

- For any given line, if we
know what is its average flow or capacity we can very well compute the mean
packet delay of the line using the queuing theory.

**- This is the basic idea behind the implementation of this algorithm.**
- This idea reduces the size of the
problem i.e., only the minimum average delay has to be calculated for the
sub net and nothing else.

- Thus, the flow based routing considers the load
and topology of the network while other routing algorithms do not.

- In few
networks, the mean data flow existing between two nodes might be predictable as
well as relatively stable.

- There occur such conditions under which the average
traffic between the two points is known.

- In such conditions the mathematical
analyzation of the flow is possible.

- This calculation can be used in the
optimization of the routing protocol.

- The flow weighted average can be
straightforward calculated which in turn can be used for the calculation of the
mean packet delay of the entire sub-net.

__The flow based routing algorithm requires the following things in advance:__
Ø Topology
of the subnet

Ø Traffic matrix

Ø Capacity
matrix

Ø A routing
algorithm

- Information flow based routing algorithms are
commonly used in the wireless sensor networks.

- These days, the measure of
information is being used a criterion for the analyzation of the performance of
the flow based routing algorithms.

- One research has put forward an argument
stating that since the sensor network is driven by the objective of the
estimation of a 2D random field, the information flow must be maximized over
the entire field and the sensor’s lifetime.

__In response to this algorithm two types of flow based routing algorithm have been designed namely:__- Maximum information routing
(MIR) and
- Conditional maximum
information routing (CMIR)

- Both of these algorithms have proved to be quite
significant when compared to the exiting algorithm – maximum residual energy
path or MREP.

## About MREP Algorithm

- This proves to be quite effective
in conservation of the energy.

- The battery energy which is limited is taken as
the most important resource. - For the maximization of the lifetime, the energy
consumption has to be balanced throughout the nodes.

- This should be done in
proportion to the resource reserves.

- This is better than routing for the
minimization of the absolute consumed power.

## About MIR Algorithm

- The ideology behind the MIR
algorithm is that there is inequality between the nodes.

- For example, two very
close nodes might not provide twice as much information provided by a lonely
node.

- Therefore, the nodes that provide more information are only given
preference.

- An additional penalty according to the node’s contribution is added
to the node for achieving the above mentioned preference.

- Dijkstra’s
algorithm is used for the computation of the shortest path.

- This helps in
sending the data to the sensor as per both the information of the origin and
the power consumed.

## About CMIR Algorithm

- This one is a hybrid algorithm and
makes use of MIR to some extent and then uses MREP algorithm for the rest of
the cycle.

- This hybrid version is better than the above two standalone
algorithms since it runs better.

## No comments:

Post a Comment