58
Chapter 3
A Frame work based on the efficient power aware
routing in MANETS
3.1 Introduction:
The nodes in an ad hoc network are constrained by battery power for
their operation. To route a packet from a source to a destination
involves a number of intermediate nodes. Hence, battery power of a
node is a precious resource that must be used efficiently in order to
avoid early termination of a node or the connected network. Hence,
energy management is an important task in such wireless networks.
Efficient battery arrangement and power management in the network
are the major issues for increasing the life of a node. These schemes
deal in the management of energy resources by controlling the early
draining of the battery power, these methods decide the proper power
level of a node and incorporate the power conservation strategies into
the protocols layers of the stack. There are the major issues and
research solutions have to be evaluated to witnesses the need of
energy management in ad hoc wireless networks.
MANET is group of mobile nodes that form a network
independently of any centralized controlled administration. Since
those mobile nodes are battery powered and extending the battery
lifetime for more survival period has become an important aim.
Most of the researchers are using different power-aware strategies
59
for efficient routing in protocols for MANETs. The mobile node in a
MANETs performs the routing function for establishing
communication among different nodes, early death of one node or
collection of nodes due to power exhaustion might cause
disconnect of services in the entire MANETs. These nodes suffer
from limited energy level problems and also as the node moves out
of the radio range of the other node, the link between them is
broken. In such cases there are two following reasons for the
linkage break Node completely dies off to energy depletion.
Node moves out of network radio range out of the
neighbourhood.
The creation of efficient routing protocols is a basic problem in a
MANET. Many different protocols are suggested in the literature [22-
25] by Wei Luo and others, each based on different situational
characters and properties. Ad-hoc networks are multiple-hop
networks, here entire nodes coordinate in managing connectivity. This
type of networks are important, where provisional network
connectivity is needed, such as in emergency or rescue operations.
Recent implementations and research have mentioned the essence of
redesigning protocols for longer battery life.
60
Much of this useful energy is wasted in buffering and listening
other transmissions. The life of the network reviewed by authors [18-
20] is usually defined according to the following criteria:
The time vested by the first node to burn out its battery power
or energy.
The time duration for the certain percentage of nodes fails to
function.
The time when the network starts getting partitioned
This issue of node failure yields into network partitioning is a
de-merit in ad-hoc networks. For designing energy efficient routing
mechanism. For this mentioned need many efforts are made to
increase the life of battery by harvesting. In this proposed section we
discuss the parameter metrics used in for energy efficient routing and
framework based on the popular energy routing protocols used
3.2 PROPOSED EPAR SIMULATION FRAMEWORK:
EPAR method provides route fairness and improves the energy
efficiency of MANETS where nodes can find the energy efficient routes
between the source and destination depending on the traffic load and
residual energy thus extending the node survival and network life
span. The dynamic structure of MANETS makes the nodes move freely
which leads to the breakage intermediate links more often make the
selected route invalid.
61
When selecting a path, the traditional DSR implementation
chooses paths with the lesser number of hops. Where in this proposed
scheme based on EPAR the path is selected based on energy. This
method calculates the energy power level of the route path, The path
selection will have lesser hop count and required energy in the path
to deliver the packet to the destination without termination due to
early depletion of nodes energy.
This can be explained on the scenarios of two paths chosen as
follows. The first route path has three hops with energy mentioned as
21, 18, and 85 joules and the second route path has four routing
hops with energy values 30, 25, 50, and 95 joules. The selections of
the route is made with reference to performance requirements, which
are adaptable for the existing ad-hoc routing protocols in use. Hence
the efficient energy aware routing schemes are inter transferable from
layer to other using adaptive ad-hoc routing protocol with the
observed merits valid.
Here in the above the energy of the first path is 18J, while that
for the second path is [Link] this factor of 25 is more than 18, the
second route path is chosen. For the figure in 3.1 the proposed EPAR
method is the on-demand protocol based on the existence and life
time prediction of the battery power. The traditional classical DSR routing
selects the shortest path AEFD or AECD but this new method selects ABCD only as
this selected route path has the maximum lifetime of existence of (1000s). The
62
strategic objective of this routing method is to extend the survival lifetime of
network with dynamic structure this method favors the path whose lifetime is
maximum
Figure3.1: Route discovery in EPAR
3.3 EPAR Mechanism:
Nodes can be either the source destination pair or any
intermediate nodes which are accountable for relaying and sending
the packets to the neighborhood network .this cooperative ability
which is inherent makes the MANETS applications use in situations of
emergency when communications are lost and quick setting of
network is needed this self healing nature of MANETS makes it use in
rescue or military applications.
63
Energy consumption of the battery and throughput are major
issues in this dynamic topology, some nodes which are intermediate
will experience more traffic load than others which will consume
more energy from their battery reserve thus depleting and lead to
early failure of the node or partitioning of the network inducing
inefficiency in the route reliability
This proposed work is on the efficient nature of the DSR routing
of hop by hop requests for new routing when demanded on reactive
basis. Majority of the wireless ad-hoc networks use route path length
metric based on the shortest path method for rote deployment
between the source, destination nodes in the network. This proposed
method of EPAR improves the QoS and route handling mechanism
.the traditional routing protocols do not have efficient routing
mechanisms based on the energy, traffic or the topological variations
When we compare DSR to EPAR, every node tracks its current
energy level using energy units. Energy field in the packet path is
updated when a node receives a route reply or ACK, Energy table is
updated with energy values in the path once the source node receives
either a route reply or an ACK. When choosing a path, the DSR
implementation chooses the path with the minimum number of hops.
However, the path is chosen based on energy for EPAR.
To enhance the performance changes have been induced in
modifying the route section and route discovery process where the
64
frame fields consists of evaluation metrics for inducing awareness in
the packets for request and reply method, her paths are selected
between the source , destination to overcome the depletion of battery
and balance traffic in different nodes.
3.3.1 Proposed Efficient Power Aware Modeling EPAR:
Algorithm Description:
Tot Energy is sum of energy levels of all i route nodes of the Route
TEnergy=∑𝑁
𝑖=1 𝐸𝑖 ……………………. (1)
The Energy each node Ei is calculated.
Ei= Pi * T…………. …………………..(2)
Consumed energy can be measured using the transmitting power or
receiving power multiply time of transmission T.
T= 8*𝑆𝑖𝑧𝑒 of packet/𝐵𝑎𝑛𝑑𝑤𝑖𝑑𝑡 assigned ……………..(3)
The Energy transmitted Eitx is defined as
Eitx = Pitx * T ……………………...…(4)
The Energy Received Eirx is defined as
Eirx = Pirx * T…………………………(5)
Consumed Energy of a node after time t is calculated using as
Eicon(t) = (Nt .C1) + (Nr .C2) ………………..(6)
Where, Eicon(t) is the energy consumed after in time t.
Nt, no. of successful packets transmitted by the node after time t.
Nr , no. of received packets by the node after time t.
C1 and C2 are constants values defined between 0 and 1.
65
The residual energy ERes of a node at time t is evaluated as
ERes = TEnergy – Eicon(t)………….. (7)
Where ERes : residual energy,
Econ : Total consumed energy of by all the Nodes is measured by
summing all node’s residual energy with the product of initial energy
and number of nodes.
TEcon = N * sum of Initial Energy – Eres…………… (8)
TEcon : total consumed energy
Route parameter for efficient route power Pi= TEcon/(N*Ei)…….(9)
The next proceeding algorithm discusses the Route Discovery
mechanism.
3.3.2 Route Discovery steps:
Step 1: The Source node be S.
Generates and sends RREQ packet with the field set as Source
Address (SA)=S, Destination Address (DA)= D, Sequence
Number= I, Then Time to Live (TTL)= T, Hops=H, BW=BWth,
Minimum energy = Initial energy;
Broadcasting the RREQ to next adjacent node if BWth >= BW.
Step 2: In case of the intermediate node receives the RREQ packet.
Then the Min energy frame field RREQ is updated with initial
energy.
Then Forward RREQ packet to next node 2.
Evaluate the Node’s Residual energy.
66
Min energy is compared with residual energy and the route
selection is based on Min energy > Res energy and BWth > BW.
Else the link between Node1 and Node2 are unavailable.
Step 3: In case the node receiving the RREQ packet in D, then the
node D.
D replies RREP packet back in a unicast manner to source.
This bandwidth of the RREP packet field is updated with the
cumulative bandwidth of the path
Also the Energy field is updated using the cumulative Energy
total.
D uni-casts all the route paths that are disjoint back to
source S.
3.3.3 Route Selection:
When the neighboring node receive RREQ packet, it will reply
with a RREP packet back to the source originated, else it
rebroadcasts the RREQ. The routes in classical DSR are selected on
the basis with the least number of hops. But the proposed EPAR
method selects the best efficient path by strategic sorting of routes
with energy efficient parameter ‘ Pi ’ in the descending manner of
nodes residual left out energy and bandwidth. This scheme forwards
the data packets based on the node route with high residual energy.
3.3.4 Route Maintenance:The link is broken if in case when the
energy values are lesser than the threshold minimum. The route
breakage is indicated by a Route Error message (RERR) which will be
67
sent back to the last sent node. Then the source node on this reply
searches and starts route new discovery procedure on receiving the
error RERR message.
3.3.5 Route Discovery:
While discovering the process, using the route request and
route reply query process cycle. The EPAR builds a route between
source to destination. The RREQ messages are broadcasted to all the
nodes in the neighborhood and these are transmitted back using
flooding methodology till they reach the destination. The fields in the
RREQ frame consists of Source, Destination, Hop Limit), Target,
Origin, energy metrics for awareness. When the RREQs finally reach
the destination or t nodes in the intermediate stage they reply with
RREP message where both have similar formats which may accept or
reject the request based on the load and the traffic situation.
When a source node wants to send a packet to destination for
which it does not already have a route, it forward a RREQ packet to all
the neighbours across the network. EPAR performance is improved by
adding energy model parameters in RREQ packet, two additional fields
are added in the RREQ header information such as bandwidth and
energy constraints. The Proposed extended format of Route Request
packet of EPAR is shown in the following Figure.3.2.
Figure.3.2: EPAR Route request packet format
68
Figure 3.3 Flow diagram for EPAR implementation
69
3.4 EPAR Packet Format:
Here the Pt value will be the power values that the packet is
transmitted on the link. Mean while for a reason a node chooses to
change the transmit power value for hop i, then it must set the Pt
value in minimum mode of transmission power to the actual
transmit power values. If the new power values differ by more than
the threshold then this corresponding Link Flag will be set.
This below table shows the data packet frame format used in
EPAR. The packet also includes the DSR fields along with the new
fields of EPAR.
IP DSR DSR Source DSR EPAR Link Data
Header Fixed Header Source Source Flag
Header Routing Routing
Address MTP
[1……N] [1……N]
Figure 3.4 EPAR data packet format
3.5 Simulation parameters:
3.5.1 Route evaluation with required Power:
In the first set of experimentation, we initially to every node give
a battery with full capacity energy or power. Using this method, no
nodes turn off due to shortage of energy.
3.5.2 Varying number of mobile nodes:
From simulation analysis, we vary the number of nodes while
maintaining the throughput, packet delivery ratio and end to end
delay constant in the network.
70
3.5.3 Varying pause time:
This section explores the effect of variable pause time over the
basic method. We execute simulations by varying the pause times in
terms of 10, 20, 30, 40 and 50 seconds for obtaining a scenario from
continuous mobility of the nodes.
3.5.4 Throughput:
This is defined as the ratio of the total amount of data packets
that reach the receiver successfully from a sender to the time taken by
the receiver to get the last possible packet is called throughput.
3.5.5. Packet Delivery Ratio (PDR):
Let P be the fraction of successful delivered packets, N be the
total number of flows, f be the identity, R be the packets received
from f and T be the transmission from f, then F can be evaluated as
3.5.6 Average End-to-End delay:
It denotes how efficient the underlying routing algorithm.
Here S will be number of packets received successfully, ri be the time
at which packets are received and Si is time at which they are sent, i
will be the unique packet identifier.
71
3.5.7 Power consumption:
In EPAR routing discovery process, the source node in the
network sends the extended RREQ message to the destination node
through number of intermediate nodes. In wireless networks the data
transmission can be directly within done with one hop or through
number of intermediate nodes. The extended RREQ message contains
the source and destination node IP address, Advertised hop count
value, Timeout value, Bandwidth of the link and minimum energy
value. The computed bandwidth and minimal nodal energy is greater
than the threshold value of bandwidth and energy then only the RREQ
message forward to the next neighbor node otherwise it discarded.
When the RREQ message arrive at next node, the bandwidth and
minimal nodal energy is updated into the route list entries. At the
initial stage the source node’s initial energy is entered into minimum
energy field, the residual energy is computed at every node in the
network.
This residual energy is compared with minimum energy field of
RREQ packet. It replaced by residual energy if this value is less than
the minimum energy field. The minimum energy should be kept as the
lowest among all the nodes in this route while selecting the best path.
The node will produce RREP packet and send back to the source node
once the RREQ packet is received by the destination node. RREP
packet is included in two additional fields Bandwidth and minimum
energy, the routing information from the source to destination is
72
recorded by the RREP packet. The duplicate packet ID is received by
the destination node, and then it responds with a maximum of RREP
packets to the source node.
When dealing with data packets over the wireless network we
believe that delay should be given the highest priority. On the other
hand, many researchers have focused and emphasized on saving
power of the node battery to last for longer time (without recharging)
and a lot of researchers. In this work, we also minimize power to an
extent that it does not degrade improved delay performance.
3.6 Results and Discussion
We have used NS-2.33 for conducting extensive simulations.
The simulation network consists of 120 nodes randomly scattered in a
2000x2000m area at the start of the simulation. Table 3.1 represents
the parameters used for simulation and for protocol evaluation. These
scripts are generated using the .tcl for the mentioned parameters.
Parameter Value
Nodes 10,20,30,40,50
Routing Protocol DSR,EPAR
Area size 1000x1000
Channel Capacity 2Mbps
Traffic Type TCP/FTP, UDP/CBR
Initial Energy 50J
73
Transmitting power 0.8J
Receiving Power 0.7J
Idle Power 0.01J
Simulation time 50 seconds
Antenna type Omni Directional
Table 3.1: simulation parameter
Power Consumption:
Due Transmissions and receptions of data packets the mobile
node battery power is consumed largely. Energy is consumed more a
node is active, and during the inactivity the node still participates in
buffering and the battery keeps discharging.
74
Figure.3.5: Energy consumption Vs Simulation time
The graph between energy consumption and simulation time is
evaluated. The graph 3.5 shows that energy consumption of DSR is
more than proposed EPAR method. Using DSR algorithm at the end of
the simulation whole of the given initial energy given at the node is
consumed making the node die and makes the network portioned.
Thus Proposed EPAR method achieves 30% power and energy saving
than the traditional DSR method.
75
Throughput:
Throughput is defined as number of successful transmission
deliveries in a given time with respect to the packets generated. We
have taken the channel capacity of 2Mbps. So maximum throughput
achieved should be 2Mbps.
Figure.3.6: Throughput Vs Simulation time
This graph depicts the performance of throughput and
simulation time, This shows that proposed EPAR achieves 20% more
throughput than DSR in the total time of simulation.
76
Network lifetime:
Network lifetime is the duration period from the deployment to
the instant when the network gets partitioned or in not functional to
whatever application it has been designated.
Figure.3.7: Network lifetime Vs Number of Nodes
A graph is plotted between network lifetime and number of
nodes. The graph shows that the lifetime of nodes in DSR routing is
drastically decreasing by increasing in the number nodes. EPAR
achieves 30% more lifetime than DSR
77
End to End Delay:
Figure.3.8: End to end delay Vs speed
End-to-end delay or One-way delay gives the time taken for
a packet to be transmitted over a system from source to
[Link] speed of network using proposed EPAR increases
significantly more when compare to DSR methods.
78
Residual Energy: Energy remaining or Residual Energy is defined as
the remaining energy of the network at the end of given transmission.
Here we have given 50J as initial energy and run simulation.
Figure.3.9: Residual power Vs Simulation time
From figure 3.9 we can observe that at end of transmission time
for EPAR (green line) has 25Joules of energy remained while for DSR
(violet) only 8 Joules remained. Thus we achieve an Efficiency of 34%
higher when compared to the traditional DSR protocol.
79
Dead Nodes:
The network lifetime and dead nodes are defined as:
1. Defined as the time taken for some% of the nodes in a network to die.
2. The time taken for the first node to die.
3. The time for all the set nodes in the network to die.
Here the graph below gives the comparison of Network lifetime of DSR
and EPAR. A node becomes dead when it runs out of its node battery
power.
Figure.3.10: Number of dead nodes Vs simulation time
80
From this Figure.3.10, we can observe that as with the
simulation time prolongs the number of dead nodes drastically
increased in DSR. While in EPAR this value is lower to the time
variation. We achieved 20% efficiency in decreasing number of dead
nodes by our proposed EPAR
Average Delay
Delay is the amount of time taken by the packet to reach from
source to destination during data [Link] in the Figure:3.11
below both DSR and proposed EPAR protocols exhibit almost the
same amount of delay. As the simulation progresses further with the
increase in the number of nodes, delay of EPAR has better
performance
Figure.3.11: Average Delay Vs number of node
81
Packet Delivery Ratio:
This is defined as the ratio of the data packets delivered to the
destination to those generated by CBR sources. The completeness of
the routing protocol are determined by this performance measure. For
effective protocol the packet delivery ratio must me high and in ideal
case it should be one.
Figure .3.12 Packet delivery ratio Vs pause time
From Figure.3.12, we can observe that up to pause time 10
seconds DSR increased up to 40% and after that increased gradually
and for pause time 50s EPAR achieved 30% more efficiency than DSR.
82
Conclusions:
Due to the Transmissions and receptions of data packets the
mobile node battery power is consumed largely though needed or not
needed. Energy is consumed more when a node is active, and during
the inactivity of the node still participates and using this proposed
method
we achieve 30% power and energy saving , 20% more
throughput than the DSR algorithm. There is a clear depiction for
improvement in terms of end to end delay and packet delivery ration
by 20 %, and to the end the residual energy is conserved for future
operations making the life time of the node extended noticeably more
in this frame work using EPAR scheme.
This method does not consider the effective traffic and
conditions of load on the traffic conditions where contending of nodes
and collusions are not taken in to account, hence this method does
not look in to these issues we propose a Restricted access (RAW)
scheme which accounts for traffic and node contention which
efficiently reduces collusions and simultaneous occupancy and
competition among the mobile ad-hoc nodes.