ISSN 2278-3091

Mohamed Najmus Saqhib et al., International Journal ofVolume

Advanced9, No.4,in July
Trends – August
Computer Science2020
and Engineering, 9(4), July – August 2020, 5691 – 5699
International Journal of Advanced Trends in Computer Science and Engineering
Available Online at http://www.warse.org/IJATCSE/static/pdf/file/ijatcse221942020.pdf
https://doi.org/10.30534/ijatcse/2020/221942020

Performance Anatomization of Routing Protocols in Wireless

Sensor Network
Mohamed Najmus Saqhib1, Lakshmikanth .S2
1
Visvesvaraya Technological University, Research center VKIT Bangalore, India, [email protected]
2
Acharya Institute of Technology, India, [email protected]


WSN consists of a set of nodes which are spread in a given
ABSTRACT area of x*y meters. Each node is defined with a unique value
Wireless Sensor Network (WSN) is used in multifarious of (xi, yi). xi is the ith position for the node, yi is the y position
applications like environment monitor, battle based systems, of the node in a two-dimensional space[9]. Each of the nodes
enemy vehicle track determination and many more. It is also is also identified with a unique id representing itself. Figure 1
limited by various constraints like cost, bandwidth, and shows the node placement strategy for the set of 100 nodes
energy consumption patterns along with network lifetime. spread across 100*100 area. The node ids are varied from
When the data packets have to be sent to the destination node Node1 to Node100.
or control center after detection, the path is established
between the detected node and the destination node [1]. When
the number of paths is more and nodes repeatedly participate
in those paths then residual energy value is also reduced of the
specific nodes which lead to holes in the network and reduces
the network lifetime. This paper presents an overview of
WSN, Lifetime ratio effects, a numerical survey of the
energy-efficient routing protocol. The methods namely
Destination Sequence Distance Vector (DSDV), Ad hoc
On-Demand Distance Vector (AODV), Zone Routing
Protocol (ZRP) and Energy Efficient Distance Routing
(EEDR) are discussed in detail along with the
implementation of these methods in MATLAB. Comparison
is performed in terms of various parameters namely delay,
hops, energy consumption, alive nodes, dead nodes, lifetime
ratio, overhead ratio, residual energy as well as throughput
[10] and it is proved that EEDR algorithm works in an
Figure 1: Node Placement Strate gy
optimized fashion.
As shown in Figure.1, the following are the positions of a few
Key words: AODV, DSDV, EEDR, Energy Efficiency, of the nodes among the 100 nodes in the network.
Lifetime Ratio, Wireless Sensor Network.
Table 1: Node position in the network
1. INTRODUCTION Node ID X Position of Y Position of the
the Node Node
Wireless Sensor Network (WSN) is a micro service based 11 9 67
system which can be used for applications varying from 9 7 11
commercial and industrial data. The features of WSN 37 42 4
includes a processor, communication protocols, amount of 31 39 65
power used and the path between the two nodes or between the 84 22 27
base station and the node [2]. Each node can perform various
tasks ranging from detecting temperature, humidity data and One of the major challenges is maintaining the [4], [5]
pressure data based transmissions between two endpoints. lifetime ratio for the network. If IE is the initial energy of the
The WSN can be used in a wide variety of services namely network then a node which has the threshold satisfying the
Military, Industry, Health Care and many more [6]. equation as below is calculated.
(1)

5691
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

And the set of nodes who do not satisfy the equation are taken (5)
in the denominator to obtain the lifetime ratio.
(2) Where UE is updated energy level [7], CE is the current
energy level and Ec is the energy consumed. For instance, if
If there are no nodes which are having the remaining energy the energy required is assumed as 20J, energy for
(RE) below IE/4 then LR will be moving towards infinity and amplification is 10J, the distance between two nodes is 10.34
vice versa. When the nodes take part in the data portage their m and environment factor is 0.7 with the node initial energy
consumed energy depends on the various factors like value as 3000J then the updated energy level can be computed
as below
attenuation factor (att), the energy required for amplification
(Eamp), the energy required for data packet transmission
UE=3000-(2*20+10*10.34^0.7) =2908.703J (6)
(Etxn) along with the distance between the nodes in the
network [8].
The remaining part of the paper is organized as follows first
(3)
the existing work present in the literature is presented,
Each time a node will be used in the path the energy level for secondly, the EEDR method is discussed in detail, thirdly
the nodes will be reduced by [20] several existing methods namely DSDV, AODV and ZRP are
discussed. The final section determines the results of the
(4) various algorithms and comparison between the algorithms

Where CE is the current energy of the node during the 2. ROUTING PROTOCOLS
computation of RE. If the value of energy required for
transmission, amplification and attenuation factor is kept The several existing methods [10] which are present in the
constant then energy variation can be plotted as in Figure 2. literature are described in this section namely AODV, DSDV
and ZONE based routing method.

2.1 DSDV routing protocol

The table used in the DSDV method [16], will maintain the
path information along with node information with respect to
unique destinations. For every period of Time 'T,' the routing
information gets updated. First, from the detection point, the
nodes which are occurring inside the transmission scope are
found out and then the path is determined from each detection
point to control center.

The time taken is found for each of the paths and the path
which has the lowest time taken is chosen for packet delivery.
Consider a set of nodes which are arranged in the format of a
Figure 2: Energy distance dependence
line in which each node is separated at a distance of 10m.
Figure 3 shows residual energy reduction for a specific node
whose initial energy level is 3000J is reduced with the number
1 -- 2 - 3- 4-5-6-7-8
of time the node participates. For the 1st time, it is 3000J and
at the end of 10 iterations, the residual energy is 2995J.
If the initiator is Node1 and the transmission range is 20m
then the set of nodes which form the cover set of Node1 is
{Node2, Node3} and if the initiator is Node 3 then set of cover
set nodes are {Node1, Node2, Node4 and Node5}. The
initiators in the DSDV method are found out using the
algorithm described in Figure 4.

The multiple paths from the initiator nodes to the destination

node are found out using the following algorithm as described
in Figure 5.

The individual path is found out using the algorithm

Figure 3: Residual Energy Reduction with the participation ratio described in Figure 6.
The reduction in energy happens using the following logic
Algorithm: Initiator Set
Input- SN and Transmission Range
5692 Description
1. SN, DN, Transmission range
2. Find the nodes in transmission range with respect to SN
N={n1,n2,……, ncs} where ni represents the ith node
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

by the following formula

Algorithm: Initiator Set
Input- SN and Transmission Range
Description
7. SN, DN, Transmission range The path which corresponds to tDSDVn will be used to
8. Find the nodes in transmission range with respect to SN
N={n1,n2,……, ncs} where ni represents the ith node deliver the data packets.
9. Measure the length of N
10. Initialize the initiator set {IS}
11. For i=1;i<=N;i++
2.3 AODV routing protocol
d. Pick the ith node ni
e. Find the nodes which are within the transmission In AODV the path maintenance overhead is reduced since
range of ni call it as TRi
f. Store the Tri into {IS}
every node does not maintain the cache of route [13]. From all
12. The final set will contain the nodes in the network, a set of nodes are chosen for storage
{TR1,TR2,…….,TRn} of route and route initiation is done only if it is required [17].
The AODV is built by modifying the DSDV algorithm. The
number of initiator nodes in the AODV is less compared to
Figure 4: Initiator Nodes Determination DSDV DSDV algorithm. After all the paths are found out the path
which has the lowest end to end distance is chosen as the best
Algorithm: Multiple Path Determination path.
Input - Initiator Nodes, DN, Transmission Range
Algorithm
1. Find the number of initiator nodes (Nl) The initiator nodes are found out by making use of the
2. For i=1,i<=Nl;i++ following algorithm as described in Figure 7.
a. Start the Timer
b. Execute the IPM method to find the path with ith node
acting as SN Algorithm: Initiator Set for AODV
c. Stop the Timer Input- SN and Transmission Range
d. Find the time difference
e. Store the Map { Ti , Pathi,} 1. SN, DN, Transmission range
f. Store the time Ti in List 2. Find the nodes in transmission range with respect to SN
N= {n1, n2… ncs} where ni represents the ith node
Output – List and Map 3. Measure the length of N
List has all the time taken across the routes 4. Initialize the initiator set {IS} as N
Map has the path and the respective time taken

Figure 7: Initiator Nodes Formation AODV.

Figure 5: Multi-Path Determination
The multiple-path determination for the AODV algorithm
can be determined by making use of the Multipath Routing
Algorithm: Individual Path Method (IPM)
Input AODV method as described in figure 8.
SN, DN and Transmission Range
Description Algorithm: Multiple Path Determination AODV
1. SN, DN and Transmission Range Input - Initiator Nodes, DN, Transmission Range
2. Find the nodes which belong to transmission range acting as
Cover Set (CS) 3. Find the number of initiator nodes (Nl)
4. For i=1,i<=Nl;i++
3. If the DN belongs to CS, then Stop
a. Execute the IPM method to find the path with ith node acting as
4. If DN is not present in CS then find the subset of CS as SN
Forward Nodes (FN) b. Compute the distance vector for the path
5. The rules are generated from the FN c. Store the Map { DVi , Pathi,}
6. The node will be picked based on rule g. Store the distance DVi in List
7. Process is repeated until path is completed Output – List and Map
List has all the time taken across the routes
Output: Path between SN and DN Map has the path and the respective time taken

Figure 8: Multiple Path Determination AODV.

Figure 6: Individual Path Determination DSDV
The objective function for determining the path sequence for
The multiple paths are found and then the cache of route data transmission is defined as below
discovery time is maintained in the format of {tDSDV1,
tDSDV2… tDSDVn}. Where tDSDV1 is the time taken to (8)
find the path1 using DSDV algorithm, tDSDVn is the time
taken to find the path n using DSDV algorithm. The best path Where DVpi is the distance on the path pi. After finding the
is found out by making use of a minimization principle used minimum distance vector then the best path is found out

5693
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

2.3. Zone Routing Protocol (ZRP) by making use of the signal to noise interference ratio (SNIR)
on the block-based transmission packets send from the source
The border nodes are found out from the neighbor nodes [18]. node. The rate of transmission and the kind of modulation
The path is found out from the border nodes using individual technique is used to find the channel state information. The
zone routing path. From the set of multiple routes route with Channel Quality Indicator (CQI) is found using the following
best time value is found out. The individual path zone routing formula
is found out as described in Figure 9

Algorithm: Individual Path Method (IPM) ZRP

Input - SN, DN and Transmission Range
-16<SNIR<14
1. SN, DN and Transmission Range
2. Find the nodes which belong to transmission Where
range acting as Cover Set (CS) CQI = Channel Quality Indicator
3. If the DN belongs to CS, then Stop SNIR = Signal to Noise Interference Ratio
4. If DN is not present in CS then find the round trip
time with respect to CS nodes The SNIR can be defined as follows
5. The node will be picked based on lowest round
trip time
6. Process is repeated until path is completed
Output -Path between SN and DN using ZRP

Where
Figure 9: Zone Routing Protocol Path Discovery Routine

2.3. Energy-Efficient Distributed Receiver (EEDR)

PT = Transmitted Power
routing protocol
LP = Path Loss
W = Signal Bandwidth
The receiver node will find the CRN packets to all nodes. For
F = Noise Figure
each of the nodes in the cover set quality of the channel is
PG = Power Gain
found out.[3] The maximum value of channel quality {CQI1,
CQI2, CQIn} is found out and then the next node is found out
The shortest path algorithm will be described in Figure 11
based on the maximum value. The initiator nodes are found
out based on AODV method. The CQI based selection
happens until TTL becomes zero. Once the TTL becomes 0 Algorithm: Shortest Path Method
then the shortest path method is triggered. The individual Input
path is found out by using the algorithm as described in SN, DN and Transmission Range
Figure 10. Description
1. SN, DN, TxnC and Transmission Range
Algorithm: Individual Path Method EEDR
Input SN, DN, TxnC and Transmission Range
2. Find the nodes which belong to transmission
Description range acting as Cover Set (CS)
1. SN, DN, TxnC and Transmission Range 3. If the DN belongs to CS, then Stop
Find the nodes which belong to transmission range acting as Cover Set
(CS)
4. If DN is not present in CS then find the distance
2. If the DN belongs to CS, then Stop with respect to CS nodes
3. If DN is not present in CS then find the CQI with respect to CS 5. The node will be picked based on lowest value of
nodes distance.
4. The node will be picked based on highest value of CQI.
5. Reduce the value of TxnC by a factor of 1 6. Repeat process until DN is reached.
6. Process is repeated until either TxnC becomes 0 or path is Output Path between SN and DN using Shortest Path
established.
7. If TxnC becomes 0 shortest path is found
Output Path between SN and DN using EEDR Figure 11: Shortest Path Method
3. RESULT
Figure 10: EEDR Individual Method
This section will present the comparison results of EEDR,
The best route is found out by making use total CQI value DSDV, AODV and ZRP algorithms. The simulation input is
which is having a maximum value. The CQI will be computed defined in Table 2. The results are based on the MATLAB
programming language.
5694
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

Table 2: Node position in the network Figure 13 shows the initial battery level for all the nodes in the
network. All the 100 nodes have been initialized with a value
Parameter Name Parameter Value of 9999 mJ.
Number of Nodes 100 Number of Iterations v/s Time Taken in ms
8
Source Node 71
Destination Node 8 7 DSDV
AODV
Transmission Range 40 ZRP
6
Initial Energy For Nodes 9999 mJ EEDR

The energy required for 20 mJ

Time Taken in ms
5
transmission
4
The energy required for 10 mJ
amplification 3
Attenuation Factor 0.5
2
Threshold Count 4
Number of Iterations 25 1

0
0 5 10 15 20 25 30 35 40 45 50
Number of Iterations

Figure 14: Delay Comparison

Figure 14 shows the delay comparison of the various methods.

The X-axis the representation of the number of times all the
algorithms are executed and Y-axis is the time taken for the
entire path. EEDR has the lowest delay as compared to other
methods namely DSDV, AODV and ZRP. The EEDR
algorithm has a delay range between 0.01 to 0.02 ms. ZRP has
a delay in the range of 0.08 to 0.09ms. AODV has a delay in
the range of 1.8 to 2.5ms. DSDV has a delay in the range of
4.3 to 7.5ms.

4 4
x 10 DSDV x 10 AODV
2.75 1.1
Figure 12: Node Formulation
2.7 1.08
Hops

Hops

Figure 12 shows the Node Formulation Module Output. As

shown in the figure the nodes are spread in a 100* 100 area. 2.65 1.06

Each Node is assigned a unique ID and there are 100 nodes in

2.6 1.04
the network. Node 71 is placed at the location (1, 10), Node 3 0 20 40 60 0 20 40 60
Number of Iterations Number of Iterations
is present at the location (80, 22).
ZRP EEDR
295 73

294.5 72.5
Hops

Hops

294 72

293.5 71.5

293 71
0 20 40 60 0 20 40 60
Number of Iterations Number of Iterations

Figure 15: Hops Comparison

Figure.15 shows the hops comparison for the period of 50

iterations. As shown in Fig the DSDV algorithm has the hops
in the range of 26500 to 27200 hops which is for all the
possible paths in the network. AODV will have the hops in
the range of 10600 to 10900 hops. ZRP has the hops of around
Figure 13: Initial Battery Level for the Nodes 294 while EEDR has the lowest hops of 72 across all paths in

5695
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

the network. The performance of EEDR is optimum for hops as Figure.18 shows a comparison of the number of dead nodes.
compared to other methods as shown in the graph The nodes whose value is less than the threshold value of
x 10
6
DSDV x 10
5
AODV (9999/4) is defined as the Y-axis. The dead nodes count at the
2.36 9.6
end of 50 iterations for EEDR algorithm is 8 nodes, followed
2.34
9.4 by ZRP which has 78 dead nodes, AODV has the next number
Energy

Energy
2.32
of dead nodes with a value of 81dead nodes along with DSDV
9.2
2.3
Number of Iterations v/s Number of Dead Nodes
90
2.28 9
0 20 40 60 0 20 40 60
Number of Iterations Number of Iterations 80
4
x 10 ZRP EEDR
2.2616 5317
70 DSDV
AODV
2.2615 5316

Number of Dead Nodes

60 ZRP
Energy

Energy

EEDR
2.2614 5315 50

2.2613 5314 40
0 20 40 60 0 20 40 60
Number of Iterations Number of Iterations
30

20

Figure 16: Energy Consumption Comparison 10

0
Figure.16 shows the energy consumption comparison of the 0 5 10 15 20 25 30 35 40 45 50
Number of Iterations
various methods... As seen in the figure the highest energy
consumption exists for DSDV method in the range of 2.29 MJ Figure 18: Number of Dead Nodes
to 2.36 MJ. AODV method has the next highest energy 5
x 10 Number of Iterations v/s Residual Energy of Network
consumption in the range of 0.9 MJ to 0.95 MJ. ZRP has the 10

highest energy consumption in the range of 22.6 KJ to 22.7 9

KJ. EEDR has the least energy consumption in the range
8 DSDV
5315.2 J to 5315.8J. EEDR has the lowest energy AODV
Residual Energy of Network

consumption followed by ZRP, AODV and DSDV. 7 ZRP

EEDR
6
Number of Iterations v/s Number of Alive Nodes
100
5
90
4
80
3
Number of Alive Nodes

70
2
60 DSDV
AODV 1
ZRP 0 5 10 15 20 25 30 35 40 45 50
50
Number of Iterations
EEDR
40
Figure 19: Residual Energy for Nodes
30 Figure.19 shows the residual energy for nodes. As shown in
20
the Figure EEDR algorithm has the highest residual energy
followed by ZRP, AODV and DSDV. At the end of 50
10
0 5 10 15 20 25 30 35 40 45 50 iterations, 88kJ of remaining energy exists in the network for
Number of Iterations
EEDR algorithm, followed by ZRP which has the residual
energy value of 25 kJ. AODV and DSDV have residual
Figure 17: Number of Alive Nodes energy of around 2kJ. Also, observe the graph when the
number of iterations increases the residual energy of the
Figure.17 shows a comparison of the number of alive nodes. network decreases.
The nodes whose value is above or equal to the threshold
value of (9999/4) is defined as the y-axis. The alive nodes Figure 20 shows the comparison of packets dropped across the
count at the end of 50 iterations for EEDR algorithm is various algorithms namely EEDR, ZRP, AODV and DSDV.
defined as 94, followed by ZRP which has 22 alive nodes, At the end of 50 iterations, there is maximum packet drop of
AODV has the next number of alive nodes with a value of 19 10 packets for EEDR followed by ZRP with a maximum
nodes along with DSDV. packet drop of 20, For AODV the number of packets dropped
is 25 and for DSDV it is 35
5696
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

Number of Iterations v/s Number of Packets Dropped

35
Figure 22 shows the packet delivery ratio comparison for all
30 the 4 algorithms for a period of 50 iterations. At the end of 50
iterations, the EEDR packet delivery ratio remains up to 94%.
Number of Packets Dropped

25
For ZRP it is 85%, AODV 80% and DSDV being the worst at
20
63%.

DSDV AODV
15 54.5 22

54

Overhead

Overhead
10 DSDV
AODV 53.5 21.5

5 ZRP
53
EEDR

52.5 21
0 0 20 40 60 0 20 40 60
0 5 10 15 20 25 30 35 40 45 50
Number of Iterations Number of Iterations
Number of Iterations
ZRP EEDR
2 2

Figure 20: Number of Packet Drop

Overhead

Overhead
1 1

Number of Iterations v/s Throughput 0 0

60

-1 -1
0 20 40 60 0 20 40 60
50 Number of Iterations Number of Iterations
DSDV Figure 23: Overhead Measure
AODV
40 ZRP
EEDR Figure 23 shows the overhead comparison between the
Throughput

30 algorithms. DSDV has the highest value of 54.2 as the

overhead followed by AODV being 21. ZRP has an overhead
20 of 0.8 and least is 0.2 for EEDR algorithm.

10
4. DISCUSSION

0
0 5 10 15 20 25 30 35 40 45 50 The percentage-wise measure of parameters across the
Number of Iterations
spectrum of algorithms – EEDR, ZRP, AODV and DSDV.
Figure 21: Throughput Comparison
Total Time (ms)
9%
< 1%
Figure 21 shows the throughput comparison between EEDR,
ZRP, AODV and DSDV algorithms. As shown in the fig
EEDR has the highest value of throughput followed by ZRP,
AODV and DSDV method. Also as the number of iterations 26%

increases the value of throughput reduces across all the

algorithms
Number of Iterations v/s PDR
1 65%

0.95

DSDV
0.9 AODV DSDV AODV EEDR ZRP
ZRP
EEDR
0.85
PDR

Figure 24: Delay Comparison Percentage

0.8

0.75
Figure 24 shows the delay comparison percentage the DSDV
algorithm has 65% utilization of time hence will have very
0.7 less time performance, followed by AODV which occupies
around 26% of the total execution time. 9% is time utilized by
0.65
0 5 10 15 20 25 30 35 40 45 50 ZRP and least is EEDR which occupies the time less than 1%.
Number of Iterations

Figure 22: Packet Delivery Ratio

5697
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

Average No Of Alive Nodes

10%
Figure.27 shows the residual energy comparison. 50% of the
total network residual energy is occupied by EEDR, followed
10%
33% by ZRP with a value of 26%, AODV and DSDV have a value
of 11%.
Average Network Lifetime
4% 3%
4%

48%

DSDV AODV EEDR ZRP

Figure 25: Alive Nodes Comparison Percentage 89%

DSDV AODV EEDR ZRP

Figure.25 shows the alive nodes comparison.48% percentage

of the alive nodes are for EEDR algorithm, followed by ZRP
Figure 28: Average Network Lifetime
33%, 10% each for AODV and DSDV methods
Figure 28 shows the average network lifetime. EEDR has
Average No Of Dead Nodes occupied around 89% of average network lifetime and
17% remaining three algorithms have 4% of overall average
network lifetime.
2%
40% Average Packet Dropped

22%

42%

7%
40%

DSDV AODV EEDR ZRP

29%

Figure 26: Dead Nodes Comparison Percentage

DSDV AODV EEDR ZRP

Figure 26 shows the dead node comparison percentage.

EEDR has the least dead node percentage of 2% followed by
ZRP which is 17%, AODV and DSDV have an equal Figure 29: Average Packet Drop
distribution of 40% dead nodes
Figure 29 shows the average packet drop EEDR has 7%
Total Residual Energy (mJ)
average packet drop, ZRP has average packet drop of 22%,
11%
AODV has the average packet drop of 29% and DSDV has
11%
28% the average packet drop of 42%.

5. CONCLUSION

The routing methods namely DSDV, AODV, ZRP and EEDR

methods are described. The energy consumption analysis
along with network lifetime measure is performed. The
Simulation results for EEDR, DSDV, AODV and ZRP
50%
concerning various parameters namely delay, hops, energy
DSDV AODV EEDR ZRP
consumed, alive nodes, dead nodes, lifetime ratio, routing
overhead, residual energy and throughput are compared.
Figure 27: Total Residual Energy

5698
Mohamed Najmus Saqhib et al., International Journal of Advanced Trends in Computer Science and Engineering, 9(4), July – August 2020, 5691 – 5699

The average network lifetime in EEDR is comparatively 11. J. Matamoros and C. Antòn-Haro, Opportunistic power
higher, nearing 89%, than other routing methods. The delay allocation and sensor selection schemes for wireless
in path establishment is less in EEDR as well as ZRP with the sensor networks, IEEE Transactions on Wireless
former performing better. The EEDR also saves big on energy Communications, vol. 9, pp. 534–539, February 2010.
whereas DSDV scores are very poor. Thus we see that for all 12. Jamal N. Al-Karaki and Ahmed E. Kamal, Routing
the parameters considered, EEDR behaves in the best fashion. Techniques in Wireless sensor networks: a survey,
Further study should be carried to study the performance of IEEE Wireless Communications, pp. 6-28, 2004
these algorithms under different buffer policies. 13. Saloua Outazgui, Youssef Fakhri RE-AODV an
Enhancement of AODV Routing Protocol for
REFERENCES Wireless Sensor Networks, in Proc International
Journal of Advanced Trends in Computer Science and
1. Kun Xie, Lele Wang, Xin Wang, Gaogang Xie, Jigang Engineering, pp. 3565-3569, December 2019
Wen, "Low Cost and High Accuracy Data Gathering https://doi.org/10.30534/ijatcse/2019/138862019
in WSNs with Matrix Completion", IEEE Transactions 14. A. Norouzi, M. Dabbaghian, A. Hatamizadeh and B.
on Mobile Computing, Vol 17, PP. 1595 – 1608. 2018. Berk Ustundag, An Improved EL Gossiping Data
2. J. Matamoros and C. Antòn-Haro, Opportunistic power Distribution Technique with Emphasis on Reliability
allocation and sensor selection schemes for wireless and Resource Constraints in Wireless Sensor
sensor networks, IEEE Transactions on Wireless Network, in Proc IEEE International IEEE Conference
Communications, vol. 9, pp. 534–539, February 2010. on Electronic Computer Technology, Kuala Lumpur,
https://doi.org/10.1109/TWC.2010.02.090215 2010 pp. 179-183.
3. Pramod Mutthigarahalli Shankarappa Shankar 15. E. Fu, H. Gao, M. Fasehullah and L. Tan, A gradient
Performance analysis of EEDR routing protocol for tracking method for resource allocation base on
WSNs, IET Wireless Sensor Systems, Vol 7, pp. 21 – 26. distributed convex optimization, 2019, in Proc 3rd
February 2017, International Symposium on Autonomous Systems
4. J. Kim, X. Lin, N. B. Shroff, and P. Sinha, Minimizing (ISAS), Shanghai, 2019, pp. 41-46.
delay and maximizing lifetime for wireless sensor 16. Mahdipour, E., Rahmani, A.M, Aminian, E.
networks with anycast, IEEE/ACM Transactions on Performance Evaluation of Destination-Sequenced
Networking, vol. 18, pp. 515–528, April 2010. Distance-Vector (DSDV) Routing Protocol, in Proc
5. L. Van Hoesel, T. Nieberg, J. Wu, and P. J. M. Havinga, International Conference on Future Networks, Bangkok,
Prolonging the lifetime of wireless sensor networks by 2009.
cross-layer interaction, IEEE Wireless 17. Chakeres, I.D., Belding-Royer, E.M., AODV routing
Communications, vol. 11, pp. 78–86, December 2004. protocol implementation design, in Proc 24th
https://doi.org/10.1109/MWC.2004.1368900 International Conference Distributed Computing
6. R. Madan, S. Cui, S. Lall, and A. Goldsmith, Modeling Systems Workshops, Tokyo, 2004.
and optimization of transmission schemes in https://doi.org/10.1109/ICDCSW.2004.1284108
energy-constrained wireless sensor networks, 18. Z. Shafiq, S. A. Mahmud; G. M. Khan, A. Sayyed, H. S.
IEEE/ACM Transactions on Networking, vol. 15, Al-Raweshidy, Zone Routing Protocol: How does it
pp.1359–1372. December 2007. perform the other way round?, in Proc International
7. D. Baumann, F. Mager, M. Zimmerling and S. Trimpe, Conference on ICT Convergence (ICTC), South Korea,
Control Guided Communication: Efficient Resource 2012.
Arbitration and Allocation in Multi-Hop Wireless 19. Iranmanesh, S, Raad, R, Kwan-Wu Chin A novel
Control Systems, IEEE Control Systems Letters, vol. 4, destination-based routing protocol (DBRP) in DTNs,
pp. 127-132. Jan. 2020. in Proc International Symposium on Communications
8. Q. Zhao and M. Gurusamy, Lifetime maximization for and Information Technologies (ISCIT), Gold Coast,
connected target coverage in wireless sensor Australia,2012.
networks, IEEE/ACM Transactions on Networking, vol. 20. Ghassan Samara , Munir Al-okour , Optimal Number of
16, pp. 1378–1391. December 2008. Cluster Heads in Wireless Sensors Networks Based
9. Tajudeen O. Olasupo, Carlos E. Otero, Framework for on LEACH International Journal of Advanced Trends in
Optimizing Deployment of Wireless Sensor Computer Science and Engineering, pp. 891 – 895,
Networks, IEEE Transactions on Network and Service February 2020
Management, Vol 5, pp. 1105-1118, September 2018. https://doi.org/10.30534/ijatcse/2020/127912020
10. S. M. Adam and R. Hassan, Delay aware Reactive
Routing Protocols for QoS in MANETs: a
Review, Journal of Applied Research and Technology,
vol. 11, pp. 844-850, 2013.
https://doi.org/10.1016/S1665-6423(13)71590-6

5699