Content uploaded by Khaled Elleithy
Author content
All content in this area was uploaded by Khaled Elleithy
Content may be subject to copyright.
TERP: A Trusted and Energy Efficient Routing Protocol for Wireless Sensor Networks
(WSNs)
Marwah Almasri, Khaled Elleithy; IEEE Senior Memebr, Anas Bushang, and Remah Alshinina
Department of Computer Science and Engineering
University of Bridgeport
Bridgeport, CT 06604
maalmasr@bridgeport.edu ,
elleithy@bridgeport.edu , abushnag@bridgeport.edu , ralshini@bridgeport.edu
Abstract—Recently, Wireless Sensor Networks (WSNs) have
emerged to provide a variety of important applications with low
cost sensors. The task of the sensors is to collect data and send it
to the sink node which delivers the data to a task manager.
However, these sensors have limited power and thus limited
lifetime. Another important consideration in WSNs is the level of
security. Transmitting data from node to another can risk the
security of the data.
In this paper, we propose a novel trusted and energy efficient
routing protocol (TERP) based on the Destination Sequenced
Distance Vector Protocol (DSDV). TERP helps to increase the
security level in the network and thus avoid any malicious nodes
or untrusted nodes. It also reduces the power consumption by
using the trust factor. The higher the degree of trust, the less
encryption is used which results in less energy. Other factors such
as drop ratio, delivery ratio, average delay, and delay jitter are
analyzed along a comparison of DSDV protocol with the
proposed TERP routing protocol.
Keywords- Wireless Sensor Network (WSN); Destination
Sequenced Distance Vector (DSDV); Trust; Trusted and Energy
Efficient Routing Protocol (TERP).
I. INTRODUCTION
Wireless Sensor Networks (WSNs) have gained increased
attention in the recent years due to its various useful
applications such as healthcare field, traffic control, and
military fields [1]. WSNs contain a large number of wireless
sensor nodes that are designed to sense and collect important
data in order to process them and then send them to the sink
node [2]. A typical sensor network consists of many scattered
sensor nodes that can communicate with each other. These
scattered nodes collect data and route it to the sink node where
the later communicates through satellite or Internet with the
task manager. Sensors are designed limited supply of power
which affects the transmission [3]. Fig. 1 illustrates the
architecture of WSNs. In WSNs it is very difficult to recharge
the battery used which has a limited energy. As a result,
considering the energy consumption is a critical factor in
WSNs [4]. To take the most advantage of wireless sensor
network, the data transmitted over the network should be
delivered and routed in a safe manner to ensure the validity
and the efficiency of the network. Security is a crucial factor
in WSN where the existence of secure routing protocols
affects the WSN’s security. There are several routing
protocols that are deployed in Mobile Ad-Hoc Network and
WSNs that are used for routing the packets between wireless
devices. Wireless Routing protocols (WRP) are classified into
three categories; proactive protocols such as DSDV, on
Demand routing protocol (reactive), and Hybrid routing
(combination of proactive and on demand) [6]. On Demand
routing also known as Reactive routing is one of the important
protocol types that helps in reducing the wireless traffic by
generating path request on demand [7]. Proactive Routing is a
table driven protocol where every node maintains a table to
register the next hop entry and the number of hops needed to
reach the destination [8]. Hybrid protocols use the
characteristics of both proactive and reactive protocols to
make routing smoother and scalable. Hybrid protocols try to
overcome the deficiencies of the other two classes of routing
protocols [9].
However, when applying such protocols many attacks
could occur by malicious nodes which cause to disturb the
efficient functionality of the network. As a result, in order to
ensure that the network provides its services without any
problem, a trust based protocol helps to resolve this issue.
Based on the trust level, each node communicates with its
neighbors [5]. Trust can help to sustain the stability of the
network and enhance the communication process between
nodes. In addition, trust can be implemented to reduce the size
of the data sent from node to node and thus decrease the
needed power consumption. As the trust level at each node
increases, less encryption or cryptography is used.
In this paper, we propose a trusted and energy efficient
routing protocol (TERP). Applying the trust concept to DSDV
protocol helps to increase the security level of the network as
it will avoid any misbehaving actions and denying malicious
nodes. Trust also can be used to increase the life of the
network as it reduces the power consumption by using less
encryption with the trusted nodes.
The rest of this paper is structured as follows: section II
presents the related work. Section III, introduces the trust
concept with in-depth discussion. Section IV, proposes a new
trusted and energy efficient routing protocol (TERP) to be
deployed in WSN. Section V, discusses and analyzes the
results. Finally, section VI, offers conclusions based on
simulation results.
Figure 1. Architecture of WSNs
II.
RELATED WORK
When dealing with WSNs, security and energy
consumption are great aspects since both have direct impact
on the authenticity, integrity, and functionality of the network.
As a result, many researchers have paid attention to the need
for improving the security level of the network to avoid the
existence of misbehaving nodes. Authors in [13] have
proposed a secure and energy efficient multipath routing
protocol called SEER. SEER protocol uses the base station
that discovers several paths to the source of the data where it
finally selects one of them for communication purposes. This
protocol updates each node with the remaining energy on a
dynamic basis in order to select the appropriate path from the
multipath chosen. The advantage of such protocol is to reduce
the extra overhead needed for maintaining the route, and
extend the sensor network's lifetime due to the efficiency of
the level of energy consumption [13].
The security of information exchanged between two nodes
is a strong factor especially in military fields. SAODV is a
secure routing protocol that assures the information security as
well as the energy efficiency. This protocol is based on the
classic AODV protocol but several mechanisms are added to
deal with security issues such as AES encryption standard,
digital signature mechanism, and RSA public-key encryption
[14]. In addition, many papers have discussed the importance
of SAODV protocol and provided some enhancements. In
[15], author has enhanced the existing SAODV protocol to
deal with serious attacks from malicious nodes that are already
have been authenticated by the network. This protocol is
called SAODV-SDDO. In order to detect these malicious
nodes, a cryptographic mechanism and a reactive approach
have been used. he basically added Intrusion Detection
Mechanism (IDM) and Trust Based Mechanism (TBM) to the
SAODV protocol [15].
Moreover, the authors in [16] have discussed various
issues regarding trust management in WSNs. They came up
with a novel trust aware routing protocol that uses both direct
and indirect trust. It has monitoring component with several
trust metrics such as network-ACK, data confidentiality,
reputation validation, data integrity, and remaining energy.
The proposed routing protocol uses three control messages
which are BEACON, REPREQ and REPRES. Each node
calculates the trust value by using the trust metric and then a
weighted trust value is computed as a direct trust value which
indicates the level of trust in the network and thus avoids and
detects many attacks [16]. In [17], a new trust scheme has
been proposed which is based on cross-layer concept. This
scheme is called Trust-Based Cross-Layer Model (TCLM). It
uses the ACKs from data link layer and TCP layer in order to
promote a trust model that avoids the malicious nods and
ensures the trusted route from source to the sink [17].
Furthermore, authors in [18] have proposed a trust based
secure data aggregation approach which is called Social
Estrangement Trust Management model (SETM). They paired
Order-Preserved Encryption Scheme (OPES) with Sigmoid
trust Model. The advantages are as follow: using a secured and
trusted data aggregation model, dealing with attacks, and
providing a keyless behavioral observation. SETM is
important in WSN since it allows nodes to adjust themselves
and pay attention to their neighboring nodes and their
trustworthiness level [18].
III. T
RUST
The existing wireless routing protocols such as Destination
Sequenced Distance Vector (DSDV), Dynamic Source
Routing (DSR) and Zone Routing protocol (ZRP) are
vulnerable to be broken and compromised since they do not
have a mechanism to detect malicious nodes that misbehave
[10]. Especially, if the attack is internally and comes from a
node or more that are inside the network.
Misbehaving is not only caused by malicious nodes but
also it can be done if the buffer of a node is fully occupied by
packets and started to drop all new incoming packets [10].
Moreover, for instance if the node is running out of power, it
could drop some packets or the node itself is broken.
By applying trust, malicious or misbehaving nodes can be
avoided during the operation of routing between the sender
and receiver which results in guaranteeing that routing data or
data packets are delivered as expected. On the other hand, trust
could also decrease the size of data or packets between the
nodes. For example, if the trust factor is high, low
cryptography can take place and this leads to reducing the
amount of bandwidth.
Wireless sensor networks have a limitation on bandwidth,
security and energy [11]. Therefore, applying the concept of
trust improves these aspects which lead to a better overall
performance of the network.
A. What is trust?
Trust in real life is simply when someone (node) acts as
expected. Trust is a replacement of knowledge and also it is
built on previous experience. In addition, trust might be used
to shift risk from someone (node) to another. There are two
types of trust; direct and indirect [11].
The direct relationship is when trust taking place between
two nodes. However, the indirect relationship when node 1
trusts node 2, and node 2 trusts node 3. Therefore, node 1
should trust node 3 but this is not always working because the
limitation of the scope is an issue in the existing reputation
systems.
1
2
4
3
Sender
Receive
r
Trusted Node
Untrusted Node
In the indirect relationship, since there is no trust between
two nodes and at the same time, a third party trusts both of
them. This third party which is a node could act as guarantor
between the two nodes that have untrusted relationship
between them which means moving the risk from node to
another. But of course, the guarantor node gains some benefit
in return. For example, the node that asked for the guarantor
should act as a guarantor node itself for some time in future as
a form of payback.
B. How to calculate the trust factor?
There are several ways to calculate the trust factor and this
is based on the need of trust. Many aspects and matrices are
involved. One of the most important aspects that play a serious
role is the data value. In other words, how important is the
data that is going to be sent from the sender to the receiver.
For example node 1 wants to send location packets to node 2
and it has a trust factor of 10 for node 2.The scale used is from
0 to 10. Trust factor 10 means that node 2 is completely
trusted by node 1 and the value of data is low. So node 1 can
use no encryption as shown in Table I. However, if node 1 has
a trust factor of 5 for node 2 and the value of data is medium,
node 1 is going to use high encryption as shown in Table I.
Energy is another aspect that should be taken in account
when trust factor is calculated because it indicates how many
bits the node can send or forward before it is down. Also,
previous experience between nodes is influential in trust factor
calculation since it shows the general direction of trusting and
if the trust factor is increasing or decreasing.
IV. P
ROPOSED WORK
The Proposed work provides information of how the
analysis was achieved and how the results were calculated by
comparing the performance of the existing DSDV and the
proposed trusted and energy efficient routing protocol (TERP).
The proposed work is evaluated based on power consumption,
drop ratio, delivery ratio, average delay, and delay jitter. The
simulation scenarios are conducted by using NS-2 simulator.
Delay is the time duration that packet need to be received at
the destination. Delay can be divided into three types. The first
one is called queuing delay which is the time duration that any
packet has to wait inside a node to be transmitted. Second,
propagation delay is the time duration that the first bit of a
packet needs to be transmitted at receiver. Finally,
transmission delay is the time taken for the whole packet to be
transmitted at the receiver. The sum of the three delays is
called total delay or latency [12].
Packet loss takes place when a packet cannot reach its
destination for any reason such as network connection
problems, lack of bandwidth, link failure or human
interference. It can be a combination of these reasons [12].
Fig. 2, demonstrates a scenario where there are trusted and
untrusted nodes in the network. In this scenario, packets are
delivered using the shortest route to the destination if the
nodes in this route are trusted. Otherwise, the second best
trusted route is used. In Fig. 2, node 3 is avoided because it is
not trusted by the network regardless it has a shorter route to
the receiver compared to node 4.
A. The First Simulation Scenario
This simulation judges the power consumption of sensor
nodes such as sender, receiver and middle nodes in a wireless
network that consists of 11 nodes using CSMA/CA protocol
on its MAC layer. Packets are transmitted from N0 to N4 for
400 seconds and each node has an initial energy of 120 mw as
shown in Fig. 3. Based on what route the packets will follow,
the nodes of this route are going to use energy when they send,
receive, or forward packets as represented in Fig. 4. At the end
of the simulation, there will be an indicator (remaining energy)
that presents information of which nodes will have more
traffic compared to others as shown in Table II. Table II,
shows the power consumption level of each node using the
untrusted DSDV routing protocol. This implementation was
conducted using the DSDV and the proposed protocol TERP
which has three levels of trust: high, medium and low.
TABLE I. The relationship between data value and trust factor.
Figure 2. A scenario of trusted and untrusted nodes in the network
High Medium Low
9 & 10
Medium
Encryption
No Encryption
NO
Encryption
6,7 & 8
High Encryption
Medium
Encryption
No
Encryption
2,3,4 &
5
High Encryption High Encryption
No
Encryption
0 & 1
- - -
Figure 3. The first simulation scenario
Figure 4. Energy level for different nodes
Using different levels of trust help to compare the power
consumption of each node for TERP routing protocol versus
untrusted DSDV routing protocol. Table III shows the power
consumption when TERP routing protocol is used with low
trust level. As shown in Fig. 5, the power consumption of
untrusted DSDV protocol is a little bit higher than the power
consumption of TERP routing protocol with low trust level.
TABLE II. The energy remaining and power consumption level of each node
using the untrusted DSDV
TABLE III. The energy remaining and power consumption level of each node
using TERP with low trust level
Figure 5. Comparison of the power consumption level between the untrusted
DSDV and TERP routing protocols with low trust level
Table IV, shows the energy remaining at each node using
TERP routing protocol with a medium trust level and also
shows the power consumption percentage used at each node.
The power consumption in this case is better than using TERP
with low trust which is much better than using the untrusted
DSDV. Similarly, increasing the level of trust saves power and
reduces the consumption level at each node as shown in Fig. 6.
TERP with high trust level can reduce the power consumption
and save energy for a longer network life. As shown in Fig. 7.
TABLE IV. Energy remaining and power consumption level of each node
using TERP with medium trust level
Nodes Energy Remaining Power Consumption%
n0 48.8860 59.26%
n1 24.7283 79.39%
n2 22.0858 81.60%
n3 43.6494 63.63%
n4 67.6472 43.63%
n5 22.1591 81.53%
n6 43.6494 63.63%
n7 68.4782 42.93%
n8 22.1493 81.54%
n9 43.6494 63.63%
n10 68.3341 43.05%
0%
20%
40%
60%
80%
100%
n0
n2
n4
n6
n8
n10
Power Consumption %
Nodes
Untrusted
DSDV
TERP with
low trust
level
Nodes Energy Remaining Power Consumption%
n0 32.3371 73.05%
n1 10.6083 91.16%
n2 7.5801 93.68%
n3 32.3371 73.05%
n4 59.541 50.38%
n5 7.8786 93.43%
n6 32.3371 73.05%
n7 61.0557 49.12%
n8 7.8042 93.50%
n9 32.3371 73.05%
n10 60.4639 49.61%
Nodes Energy Remaining Power Consumption %
n0 39.6778 66.94%
n1 13.5479 88.71%
n2 10.5215 91.23%
n3 33.6805 71.93%
n4 59.2806 50.60%
n5 10.9727 90.86%
n6 33.6805 71.93%
n7 63.4389 47.13%
n8 10.5924 91.17%
n9 33.6805 71.93%
n10 61.0394 49.13%
TERP shows a better performance in terms of saving energy as
it reduces the power consumption percentage compared to the
untrusted DSDV. Table V, summarizes the energy remaining
at every node along with the percentage of power consumed.
As shown in Fig. 8, TERP with high trust level consumes
less power compared to untrusted DSDV and low trusted
TERP. Therefore, using TERP with high or medium level of
trust can defiantly save a lot of energy and enhance the life of
the whole network.
TABLE V. summarizes the energy remaining and power consumption level
of each node using TERP with a high trust level
Figure 6. Comparison between the power consumption level between the
untrusted DSDV and TERP routing protocols with a medium trust level
Figure 7. Comparison of power consumption level between untrusted DSDV
and TERP routing protocols with a high trust level
Figure 8. Comparison of the consumption of different levels of trust along
with the untrusted DSDV
B. The Second Simulation Scenario
This simulation scenario is used to examine the
performance of the number of dropped packets, dropped ratio,
delivery ratio, average end-to-end delay, and delay jitter.
There are 11 nodes that are created in this scenario. The nodes
act as sensor nodes in a wireless network that are using
CSMA/CA protocol. The sender node, which is n0, is going to
transmit packets to the receiver node n4 at time 100 seconds
and it will stop transmitting at time 500 seconds as shown in
Fig. 9. Also, node n3 at time 300 seconds starts misbehaving
by dropping packets. The number of packets sent per second is
increased to calculate the performance metrics in different
environment situations. This implementation was applied
using the existing DSDV and the proposed protocol TERP.
Results of both protocols were collected and compared to each
other to know which one is performing better.
1) Drop Ratio:
The drop ratio indicates how often packets are dropped in
the network during the simulation time. Table VI shows the
number of sent packets which are 8000, 16001, 32001, 64000,
and 128000 packets and it also summarizes the number of
received packets and dropped packets, and the drop ratio of
untrusted DSDV in each case. The drop ratio is calculated
using (1).
N D P
N S P
X 100 (1).
As shown in Table VI when sending 8000 packets, half of the
packets are received and the other half is dropped which leads
to a 50% drop ratio. As the number of sent packets increases,
the drop ratio of using untrusted DSDV increases as well. On
0%
20%
40%
60%
80%
100%
n0
n2
n4
n6
n8
n10
Power Consumption %
Nodes
Untrusted
DSDV
TERP With
Medium
Trust Level
0%
20%
40%
60%
80%
100%
n0
n2
n4
n6
n8
n10
Power Consumption %
Nodes
Untrusted
DSDV
TERP with
High Trust
Level
0%
20%
40%
60%
80%
100%
n0
n1
n2
n3
n4
n5
n6
n7
n8
n9
n10
Power Consumption %
Nodes
Untrusted DSDV
TERP With High Trust Level
TERP With Medium Trust Level
TERP With Low Trust Level
Nodes Energy Remaining Power Consumption%
n0 53.1161 55.74%
n1 29.3538 75.54%
n2 24.9656 79.20%
n3 44.9139 62.57%
n4 67.3131 43.91%
n5 26.0615 78.28%
n6 44.9139 62.57%
n7 68.592 42.84%
n8 25.7538 78.54%
n9 44.9139 62.57%
n10 68.1101 43.24%
the other hand, when sending the same number of packets
using TERP protocol, 7321 packets are received whereas 679
packets are dropped leading to 8.49% drop ratio as represented
in Table VII. Again as the number of sent packets increases,
the drop ratio increases as well. However, comparing TERP
with DSDV, TERP has much better drop ratios as it happens
to drop less number of packets as shown in Fig. 10.
2) Delivery Ratio
Delivery ratio indicates how many packets are received
compared with the number of sent packets during the
simulation. Using untrusted DSDV and TERP protocols lead
have different delivery ratios. The delivery ratio is calculated
using (2).
N R P
N S P
X 100 (2).
Table VIII provides the delivery ratios for both DSDV and
TERP routing protocols. Sending 8000 packets has a delivery
ratio of 50%, where it is 91.51% using TERP protocol. As the
number of sent packets increases, the delivery ratio decreases.
However, with TERP, the delivery ratio is much better
compared with DSDV protocol. For example, sending 16001
has resulted in 47.65% and 82.71% for DSDV and TERP
respectively. Therefore, using the proposed TERP helps to
maximize the number of received packets and thus increases
the delivery ratio as shown in Fig. 11. Fig. 11, compares the
delivery ratio of untrusted DSDV and TERP protocols where
it highlights the difference in the performance of both
protocols. TERP has a higher delivery ratio than DSDV and
reaches about the same level when sending a large number of
packets such as 128000 packets.
Figure 9. The second simulation scenario
TABLE VI: The drop ratio of untrusted DSDV routing protocol
Number
of Sent
Packets
Number of
Received
Packets
Number of
Dropped
Packets
DSDV
Drop Ratio%
8000 4000 4000 50%
16001 7624 8377 52.35%
32001 6665 25363 79.26%
64000 7656 56345 88.03%
128000 7656 120344 94.02%
TABLE VII: The drop ratio of TERP routing protocol
Number
of Sent
Packets
Number of
Received
Packets
Number of
Dropped
Packets
TERP
Drop Ratio%
8000 7321 679 8.49%
16001 13235 2766 17.29%
32001 11941 20060 62.69%
64000 12245 51755 80.87%
128000 12380 115620 90.33%
Figure 10. Comparison of the drop ratio for both untrusted DSDV and TERP
routing protocols
Table VIII provides the delivery ratios for both DSDV and
TERP routing protocols. Sending 8000 packets has a delivery
ratio of 50%, where it is 91.51% using TERP protocol. As the
number of sent packets increases, the delivery ratio decreases.
However, with TERP, the delivery ratio is much better
compared with DSDV protocol. For example, sending 16001
has resulted in 47.65% and 82.71% for DSDV and TERP
respectively. Therefore, using the proposed TERP helps to
maximize the number of received packets and thus increases
the delivery ratio as shown in Fig. 11. Fig. 11, compares the
delivery ratio of untrusted DSDV and TERP protocols where
it highlights the difference in the performance of both
protocols. TERP has a higher delivery ratio than DSDV and
reaches about the same level when sending a large number of
packets such as 128000 packets.
0%
20%
40%
60%
80%
100%
8000
16001
32001
64000
128000
Drop Ratio %
Number of Sent Packets
Untrusted
DSDV
TERP
TABLE VIII: Summary of the delivery ratio of DSDV and TERP routing
protocols
Number of Sent
Packet
DSDV Delivery
Ratio%
TERP Delivery
Ratio%
8000 50.0% 91.51%
16001 47.65% 82.71%
32001 20.83% 37.31%
64000 11.96% 19.13%
128000 5.98% 9.67%
Figure 11. Compares the delivery ratio for both untrusted DSDV and TERP
routing protocols
3) Average End-to-End Delay
Table IX gives the average delay for both DSDV and the
proposed TERP protocol. As shown in Fig. 12, the average
delay values for the TERP are higher than those for the
untrusted DSDV over an interval. When the interval is 0.05,
the average delay of DSDV is 0.0249 seconds while it is
0.0274 seconds using TERP. This is due to more information
is sent to the trusted nodes without encryption which takes
longer delays to process. However, in the case of untrusted
nodes, only relevant encrypted information is sent and hence it
is processed faster. Also as shown Fig.12, as the interval
decreases, the average delay increases as well.
4) Delay Jitter:
In order to calculate the delay jitter, the maximum and
minimum delays for a packet at each interval are needed as in
Table X. Fig. 13 shows the delay jitter for both DSDV and
TERP protocols. As expected, the trusted nodes take longer
time than the untrusted nodes because of the amount of
information sent to each node. When the interval is 0.003125
the delay jitter for DSDV is 1.4375 seconds where it is 11.668
seconds for TERP as provided in Table XI. As a result, as the
interval decreases, the delay jitter increases.
TABLE IX: Comparison of the average delay of DSDV and TERP routing
protocols
Interval
Average Delay of
DSDV
Average Delay of
TERP
0.05 0.0249 0.0274
0.025 1.0709 1.2739
0.0125 1.5989 1.6033
0.00625 1.3294 1.4685
0.003125 1.332 1.465
Figure 12. Comparison of the average delay for both untrusted DSDV and
TERP routing protocols
TABLE X: The maximum and the minimum average delays for a packet.
TABLE XI: The delay jitter of DSDV and TERP protocols.
Interval
Delay Jitter of
DSDV
Delay Jitter of
TERP
0.05 0.0222 0.0222
0.025 2.1029 2.6773
0.0125 3.7522 10.3763
0.00625 1.5185 25.9951
0.003125 1.4375 11.668
Figure 13. The jitter delay for both untrusted DSDV and TERP routing
protocols
0%
20%
40%
60%
80%
100%
8000
16001
32001
64000
128000
Delivery Ratio %
Number of Sent Packets
Untrusted
DSDV
TERP
0
0.5
1
1.5
2
Average Delay
Interval
Untrusted
DSDV
TERP
0
5
10
15
20
25
30
Delay Jitter
Interval
Untrusted
DSDV
TERP
Interval
Maximum Delay
for a Packet
Minimum Delay
for a Packet
0.05 0.0447 0.0225
0.025 2.1478 0.0448
0.0125 3.8029 0.0506
0.00625 1.9283 0.4098
0.003125 1.9283 0.4908
V. ANALYSIS OF RESULTS
Results show that TERP which is based on is performing
better than the existing DSDV. The remaining energy in nodes
after the simulation is higher than those used DSDV without
trust. In other words, trust saves more power. The main reason
why trust improves the power consumption is that it reduces or
even removes cryptography in case of high trust from
transmitted packets which results in that nodes need to
transmit fewer bits for each packet than usual. This leads to
less energy consumed in nodes. For instance, the sender node
n0 saved 17.31% of its energy in case of using high trust. Also
the receiver node n4 saved about 6.47% of its power and the
energy that being saved in middle nodes was up to 15.62% as
shown in Table II, and Table V. These results demonstrate that
trust can improve the lifetime of a sensor network.
Moreover, based on our findings, TERP is performing
much better than DSDV without trust. For example, when
node n3 starts misbehaving and dropping packets, the existing
DSDV does not have the mechanism to detect weather n3 is
misbehaving or not. Therefore, DSDV without trust will think
n3 is still acting in a good way and will continue to transmit
packets to it while n3 is misbehaving. However, the proposed
protocol TERP detects that n3 is not acting as expected. So, n3
is removed from the trusted nodes and added to the untrusted
nodes. Furthermore, the proposed solution recalculates the
route to the destination using only the route that all its nodes
are trusted. Delivery ratio can be improved up to almost the
double in the trusted scenario as shown in Fig. 11. However,
the average end-to-end delay and delay jitter are increased in
TERP due to more packets are and time duration needed to
recalculating routes to the destination that does not have
untrusted nodes as shown in Fig. 12.
VI. C
ONCLUSION
In wireless sensor networks (WSNs), saving energy is
challenging as it is hard to recharge or replace the sensors
used. This paper proposes a trusted and an energy efficient
protocol called TERP that helps to maximize the network life.
Simulation results show that TERP reduces the power
consumption compared to the existing routing protocol DSDV
using the trust concept. The more the level of trust is, the less
encryption is needed. Three levels of trust are discussed in
depth and compared to DSDV in terms of power consumption.
Other factors such as drop ratio, delivery ratio, average delay,
and delay jitter are also analyzed and discussed. TERP has less
drop ratio and more delivery ratio than DSDV.
R
EFERENCES
[1] Zahariadis, T.; Trakadas, P.; Leligou, H.; Karkazis, P.; Voliotis, S.,
"Implementing a Trust-Aware Routing Protocol in Wireless Sensor
Nodes," Developments in E-systems Engineering (DESE), 2010
pp.47,52, 6-8 Sept. 2010.
[2] García Villalba LJ, Sandoval Orozco AL, Triviño Cabrera A, Barenco
Abbas CJ. “ Routing Protocols in Wireless Sensor Networks,” Sensors.9,
No. 11, 2009.
[3] Praveena, A.; Devasena, S.; Chelvan, K.M.A., "Achieving energy
efficient and secure communication in wireless sensor networks,"
Wireless and Optical Communications Networks, IFIP International
Conference, 2006.
[4] Kulkarni, N.; Prasad, R.; Cornean, H.; Gupta, N., "Performance
Evaluation of AODV, DSDV & DSR for Quasi Random Deployment of
Sensor Nodes in Wireless Sensor Networks," Devices and
Communications (ICDeCom), 2011 International Conference on , vol.,
no., pp.1,5, 24-25 Feb. 2011.
[5] Pushpa, A.M., "Trust based secure routing in AODV routing protocol,"
Internet Multimedia Services Architecture and Applications (IMSAA),
2009 IEEE International Conference on , vol., no., pp.1,6, 9-11 Dec.
2009.
[6] Pandey, A. K., & Fujinoki, H. (2005). Study of MANET routing
protocols by GloMoSim simulator. International Journal of Network
Management , 15, 393-410.
[7] Koliousis, A., & Sventek, j. (n.d.). Proactive vs Reactive Routing for
Wireless Sensor Network. 7-8.
[8] Tyagi, S. S., & Chauhan, R. K. (2010). Performance Analysis of
Proactive and Reactive Routning Protocol for Ad hoc Networkds.
International Journal of Computer Applications , 1, 27-28.
[9] Sharma, m., & Singh, G. (2011). Evaluation of proactive, Reactive and
Hybrid Adhoc Routing. International Journal of Smart Sensors and
Ad Hoc Networks , 1 (2), 65-66.
[10] Arif, M. Z., & Shrivastava, G. (2012). Trusted Destination Sequenced
Distance Vector Routing Protocol for Mobile Ad-hoc Network.
International Journal of Computer Application , 15, 0975-887.
[11] Gordon, R. L.; Dawoud, D. S., "Direct and indirect trust establishment in
ad hoc networks by certificate distribution and verification," Wireless
Communication, Vehicular Technology, Information Theory and
Aerospace & Electronic Systems Technology, 2009. Wireless VITAE
2009. 1st International Conference on , vol., no., pp.624,629, 17-20 May
2009.
[12] Peterson, L. & Davie, B. Computer Networks Edition 4, San Francisco:
Morgan Kaufmann, 2007.
[13] N. Nasser and Y. Chen, “SEEM: secure and energy-efficient multipath
routing protocol for wireless sensor networks,” Computer
Communications, vol. 30, no. 11-12, pp. 2401–2412, 2007.
[14] Li Wei, Chen Ming, Li Mlingming. "Information Security Routing
Protocol in the WSN". The Fifth International Conference on
Information Assurance and Security. Xi’an, 2009. 651-656.
[15] F.De Rango, “Improving SAODV Protocol with Trust levels
management, IDM and Incentive Cooperation in MANET,” in Wireless
Telecommunication Symposium (WTS’09), Prague, Czech Republic, 22-
24 Apr.2009.
[16] Theodore Zahariadis, Panagiotis Trakadas, Helen Leligou, Panagiotis
Karkazis, “Implementing a Trust-Aware Routing Protocol in Wireless
Sensor Nodes”, DeSE 2010, London UK, 6-8 September 2010.
[17] Rahhal, H.A.; Ali, I.A.; Shaheen, S.I., "A novel Trust-Based Cross-
Layer Model for Wireless Sensor Networks," Radio Science Conference
(NRSC), 2011 28th National , vol., no., pp.1,10, 26-28 April 2011.
[18] Poolsappasit, N.; Busby, M.; Madria, S.K., "Trust Management of
Encrypted Data Aggregation in a Sensor Network Environment," Mobile
Data Management (MDM), 2012 IEEE 13th International Conference
on , vol., no., pp.157,166, 23-26 July 2012.
