Energy Efficient Data Gathering Algorithms in Sensor Networks

coachkentuckyAI and Robotics

Nov 25, 2013 (3 years and 6 months ago)

73 views

Vikramaditya

What is a Sensor Network?


Sensor networks

mainly constitute of
inexpensive sensors densely deployed
for data collection from the field in a
variety of scenarios


A sensor node is an autonomous device
with integrated sensing, processing, and
communication capabilities


Data gathering in Sensor Network


A typical application in sensor network is
gathering sensed data at a distant base
station.


Each sensor node has power control and
the ability to transmit data to any other
sensor node or directly to the BS.


In each round of this data gathering
application, all data from all nodes need to
be collected and transmitted to the BS,
where the end user can access the data


The difficulties in collect and transmit
data in Sensor Network


Each node transmits its data directly to
the Base Station? Not suggested

1.
BS is usually far away from sensors


such communication will be a high cost


and drain the power quickly

2.
Sensor's battery is not replaceable, and
sensors may operate in hostile or remote
environments



Energy consumption is considered as the
most important concern in sensor network





Approaches for Data Gathering

Direct Approach


Each sensor sends its data directly to the
base station


quickly drain the battery of the nodes and
reduce the system lifetime



the only receptions in this protocol occur at
the base station



Only the base station is close to nodes, or
the energy required to receive data is
large, direct approach

may be an
acceptable (and possibly optimal) method
of communication


MTE Based Approach


Nodes route data to the Base Station
through intermediate nodes



Intermediate nodes act as routers for other
nodes’ data and work as sensing nodes as
well



Intermediate nodes are chosen if and only
if the transmit amplifier energy

is minimized



Clustering Based Approach


Nodes are organized into groups, or say
clusters



Each cluster has a cluster head



Other nodes in the same cluster sends
data to its cluster head



Cluster heads transmit the data to the
Base Station



Only cluster heads can send data to the
Base Station


Clustering Based Approach (Continue)


Static &
Dynamic Clustering
Approach



The difference is the way to choose
cluster heads


Static
Clustering Approach

has a fixed
cluster heads


Dynamic Clustering Approach has
dynamic cluster heads

Clustering
Approach

An Example (Continue)

All nodes marked with the same symbol belong to the same

Cluster and the cluster
-
head nodes are marked with a

.

Chain Based Approaches


Each node receives from and transmits to close
neighbours and takes turns being the leader for
transmission to the Base Station



Assumed that all nodes have global knowledge of
the network and employ the greedy algorithm



Starts with the furthest node from the BS

to
ensure
that nodes farther from the Base Station have
close neighbours



Data gathering is performed in rounds. In each
round, each node receives data from one
neighbour, fuses with its own data, and transmits
to the other neighbour on the chain


Chain Based Approaches

An Example

node c(2) is the leader. Node c(0) will pass its data to node c(1).

Node c(1) fuses node c(0)’s data with its own

and then transmits to the leader.

Node c(3) and c(4) do the same thing.

Node c(2) waits to receive data from both neighbours c(1)

and c(3)

and then fuses its data with its neighbours’ data.

Finally, node c(2) transmits one message to the BS


E
NERGY
-
E
FFICIENT

D
ATA

G
ATHERING

WITH

M
ULTIPLE

P
ATHS





Multiple Path
Construction
Mechanism

Data
Forwarding
Mechanism

Multiple Path Construction
Mechanism


Data Forwarding Mechanism


AN EXAMPLE SCENARIO

Performance evaluation
Environment


NS 2 Simulation


200 Nodes


1000 m x 1000 m Area


Each simulation runs for 300 Seconds


Each Node transmission range 250 m


CBR 40 Bytes Sized Traffic


Energy at Each Node 10 J


Energy Transmitting Data 0.6W


Energy Receiving Data 0.3 W


Parameters Compared


Experiment 1



Comparison EDGM with AODV


a] Throughput


b] Nodal Life


Experiment 2



Energy Saving not considered.



Random Selection Technique was
considered.


a] Nodal Life in Dense Network,


b] Nodal Life in Sparse Network.



Throughput : EDGM vs. AODV

Nodal Life: EDGM vs. AODV

Nodal Life: Dense Network

Nodal Life: Sparse Network

Conclusion


Energy consumption is considered as
the most important concern.


It is hard to say which approach we
point in this paper is the best one. We
only choose the approach fitting for the
particular case.


Should talk more about there NS 2
Simulation should include Energy
Equations if any included.


Static and Dynamic nodes evaluation
should be included.