PSO-GSA Tuned Dynamic Allocation in Wireless Video Sensor Networks for IOT

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 04 Issue: 07 | July -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 1949
PSO-GSA Tuned Dynamic Allocation in Wireless Video Sensor Networks
for IOT
Shilpa Bisen,
M.Tech ,Department of Electronics and Communication,
AMITY University, Noida, U.P., India
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract— We are witnessing the formation of an Internet of
Things (IoT), where real-world entities (e.g., people, plants,
cars) augmented with computing devices (e.g., smart phones,
tablets, sensor no des), sensors (e.g., humidity sensors,
microphones, cameras), and actuators (e.g., motors, LED) are
connected to the Internet, enabling them to publish their
generated data on the Web. Two main characteristics of the
IoT are its large scale interconnecting billionsofSmartThings
in the next decade, and the resource limitations of Smart
Things (i.e., of their embedded computing devices). For many
IoT applications, routing and sensor search are essential
services. Our thesis is that despite the large scale of the IoT
and the resource limitations of Smart Things, efficient
solutions for the routing and sensor search services can
actually be provided. We support this thesis by proposing,
implementing, and evaluating routing algorithm for high-
scale wireless sensor networks (which are a building block of
the IoT), and channel allocation for those searched path.
Particle Swarm Optimization (PSO) and Gravitational Search
Algorithm (GSA) are used to assign the optimal channel
bandwidth to all paths out of total assigned bandwidth to
minimize the total path loss so that all data canbetransferred
to destination.
Key words: Particle swarm optimization, Gravitational
Search Algorithm.
1.INTRODUCTION
1.1 Internet of Things:
Every kind of things we think, we will be connected in
Internet of Things (IoT). In this, at any time we can get
information about anything from anywhere. This
interconnected ecosystem have so many countless
applications which interface with the internet; embedded
and resource-constrained devices will imminentlydominate
a domain once accessible exclusively to servers and
workstation computers, providing highly specific functions
for sensing and actuating. Such devices find new ways for
communicating with existing infrastructure and outside the
world and between the things which are outside of their
applications. So, we need a broad vision for the IoT for
everything we might need. Figure 1.1 shows the vision of
Internet of three layers.
The core internet is the first layer of this vision. It is the
backbone of the internet. It is made up of routers, servers
and carrier infrastructure which are machinesandaccess all
the way along the route which connect to a website and its
endpoint also. The ‘fringe’ Internet is the 2nd layer which
included the concept of workstation computers; personal
desktop and laptop computers. The internet of things is the
third and final layer which can act either clients, servers or
both and provide so many services and be any type of
devices
Figure 1.1: Internet of Things vision
1.2 Wireless Sensor Networks
Efficient design and implementation of wireless sensor
networks has become a hot area of research in recent years,
due to the vast potential of sensor networks to enable
applications that connect the physical world to the virtual
world. By networking large numbers of tiny sensor nodes, it
is possible to obtain data about physical phenomena that
was difficult or impossible to obtain in more conventional
ways. Potential applications for such large-scale wireless
sensor networks exist in a varietyoffields,includingmedical
monitoring, environmental monitoring, surveillance, home
security, military operations, and industrial machine
monitoring. Protocol design for sensor networks must
account for the properties of ad hoc networks, including the
following.

 Lifetime constraints imposed by the limited energy
supplies of the nodes in the network.
 Unreliable communication due to the wireless
medium.
 Need for self-configuration, requiring little or no
human intervention.
Figure 1.2: Wireless sensor networking
Attacks can be categorized as insider (internal)andoutsider
(external) Attacks can also be classifiedaspassiveandactive
attacks. Internal attacks: In this type of attack, nodes of the
networks perform abnormally. It can destroy the entire
network easily by using compromised node. External
attacks: In this type of attack, nodes whichperformedattack
function, are not belongs to the network. So, they have no
information such as cryptographic of the network. Passive
attacks: These attacks are outside from thenetwork sothey
do not direct affect the network. Active attacks: These
attacks are inside the network and disturb all the activity of
the network. It also interrupts theinformation,modification,
traffic analysis, and traffic monitoring.
We set our objectives for the dissertation work as:
 To allocate the channel and path to video data to
minimise the energy consumption, the non linear
programming formulation will be used. This NLP
problem will be solved by more accurate
optimisation algorithm.
 To reduce the power consumption only active
nodes (which will be triggered at some activity
detection) will be considered.
 An IOT network of 100-400 nodes will be
considered and whole proposed work will be
simulated in MATLAB environment.
2.PROPOSED WORK
This work is targeted to minimize the powerconsumptionin
wireless video application of IOT. The mathematical
formulation of this, converted the solution in to a non linear
equation which can't be solved linearly. Optimization
algorithms has to be applied on them.
The mathematical problem for this work is represented as:
Where = transmitted power and is link utilisation
ratio. The Ptx and UR will be calculated between two sensor
nodes termed as edge. The choice of optimal edgessatisfying
the above equation is subject to constraint that
I. ; where ‘e’ is the edge and ‘E’ is the set of all
alive edges
II. The remaining link energy at each edge must be to
hold the new sensor data.
This work is divided into two sections. In one part route
selection is done and in another part channel allocation to
each route is done using hybrid PSO-GSA algorithm in
channel allocation part. Let D is the distance from a source
node to a destination node. A general formula for the path
loss for a wireless link is given by (in unit of dB) :
Here is wavelength of the RF signal. The TX power PTx(e)
for each edge can then be represented as follows (in unit of
mW):
Here is a channel factor. As describedabove,themaximum
link rate is assumed as 100Mbps, which is then with
distortion factor β, a possible link rate for each edge e canbe
defined by
The total data rate traversing an edge e is defined as U(e).
Then the link utilization ratio UR(e) for edge e is defined by
Here R(e) is a possible link rate for edge e. For each edge e,
the effective TX power is defined by
Hence the total effective power consumption of the
entire network is defined as
Using these set of equations route selection between active
nodes are searched. Once route is finalized then all edges of
these paths are used to allocate channels. Each edge or path
gets a dedicated channel for transmission once it is
requested. Dynamic channel allocation is used to allocate
channels and each channel is of same bandwidth which is a
part of total bandwidth available in our work.

Dynamic Channel Allocation by PSO-GSA
In our proposed scheme PSOGSA optimized technique is
used which is hybrid of PSO (Particle Swarm optimization)
and GSA (Gravitational Search Algorithm) optimization
algorithm which requires an objective function to minimize.
Steps of proposed algorithms are described as:
Step1. Initialize all initial parameters like active
nodes, their position,channel bandwidth,frequency
to model it.
Step2. Place the nodes randomly in geographical
region of 10*10
Step3. Mark the transmission range of each node
to be 5 meter and plot a circular region aroundeach
node.
Step4. Manage a sink nodeand selectthebestpath
for each node to sink based on minimum hopes and
distance.
Step5. Pass these paths for each node to hybrid
PSO-GSA optimization algorithm to get the tuned
bandwidth allocation for minimization of power
consumption.
PSO-GSA Initialization
Step6. Initialize the random positions of particles
in PSO.
Step7. Consider the searching space dimensionas
number of available channels
Step8. Initialize the weighting parameters of PSO
as 1.2 and 0.5.
Step9. Compare the fitness value of each particle
with the previous best position of bacteria.Iffitness
function value is less for this new position than
previous position then it will be assigned as new.
Step10. The present bets position is termed as
current position of particle for PSO and output of
fitness function is local for the PSO.
GSA Starts here:
Step11. The current position selected in previous
step is used to get the mass for each agent as per
GSA algorithm. The minimum value of fitness
function is selected as best and maximum as worst
position and using the formulas, mass of each agent
can be calculated as:
Step12. Gravitational force is calculated as:
The formula is described in section 3.1.
Step13. This new velocityisthedirectionofparticle
in PSO is updated as
Here gbest is the global best position of particles in
PSO and acceleration is calculated in GSA as
.
GSA ends here
Step14. The final position of agents which is
achieved either by matching the condition ofpower
reduction or by reaching the maximum iterations.
Step15. Final positions of agents thus settled are
considered as the final weighted sequence of PTS
algorithm and multiplied with input sub blocks and
PAPR is calculated.
3.Result
Our results are tested fordifferentactivenodesnumbers and
comparison is done in the basis of effective power and path
loss component, tuned with Geneticalgorithm,GSAandPSO-
GSA. We used 6 channels of 20 MHz bandwidth as per IEEE
802.11 standard. There is no limit to multiple times use of
these channels during simulation.Soifthereismorenumber
of edges in a path then these channels can be allotted
dynamically to each edge. This channel allocation process is
optimized by PSO-GSA algorithm. Figure 3.1 shows the
random placement of 40 active nodes in a geographical
region of 10*10 square meters with transmission range of 5
meter.
-4 -2 0 2 4 6 8 10 12 14 16
-4
-2
0
2
4
6
8
10
12
14
16
1
2
345
6
7
8 9
10
11
12 13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 40
X coordinates of nodes
ycoordinatesofnodes
Nodes Placed in geographical region of 10*10
Figure 3.1: Randomly allocated active nodes in a
geographical region of 10*10 m2
The red square box denotes the sink node position and blue
hexagons are other nodes. These are all active nodes. Only
nodes in this circular area can transfer their data to centre
node. We programmatically sorted all nodes which are in
range of each node and these path are plotted in figure 3.2
for each node to each other in range node.Out ofall available
paths form each node to sink, the best one is chosen on the
basis of number of hops. Least is the number of hops, final is
the path to sink. This final path is shown in figure 3.3. A final
path table for each node to sink node is shown in table 3.3.

1 2 3 4 5 6 7 8 9 10
1
2
3
4
5
6
7
8
9
10
11
1
2
3
45
6
7
8 9
10
11
12 13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 40
X coordinates of nodes
ycoordinatesofnodes
All possible edges in geographical region of 10*10
Figure 3.2: all possible edges for each node for nodes in
figure 3.1
1 2 3 4 5 6 7 8 9 10
1
2
3
4
5
6
7
8
9
10
1
2
3
45
6
7
8 9
10
11
12 13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39 40
Final selected path for each node to sink
Figure 3.3: Finalized path for each node to sink node
Once the paths are selected then the NP hard problem of
optimal allocation of channels isleft,whichisveryimportant
to reduce the power consumption in the network. For this
we have used hybrid PSO-GSA optimization which is better
than genetic algorithm (GA) and GSA as GA is a local
optimization algorithm which sometime skips local minima
points during the search of minimum point whereas GSA is
the global optimization algorithm which checks every point
in search of global minima.
1 -> 8 -> 9
2 -> 8 -> 9
3 -> 9
4 -> 9
5 -> 9
6 -> 9
7 -> 9
8 -> 9
10 -> 9
11 -> 9
12 -> 9
13 -> 9
14 -> 9
15 -> 9
16 -> 9
17 -> 9
18 -> 9
19 -> 9
20 -> 9
21 -> 9
22 -> 8 -> 9
23 -> 9
24 -> 9
25 -> 9
26 -> 8 -> 9
27 -> 8 -> 9
28 -> 9
29 -> 9
30 -> 8 -> 9
31 -> 9
32 -> 8 -> 9
33 -> 9
34 -> 9
35 -> 9
36 -> 9
37 -> 8 -> 9
38 -> 1 -> 8 -
> 9
39 -> 9
40 -> 9
Table 3.3: Finalized path from each node to sink for nodes
in figure 3.1
After optimization the available channel bandwidth of
channels allotted by both optimizations are shown in table
3.4.
Available Channel
bandwidth
PSO-GSA 20 20 20 1 11
20
GSA 5 8 3 4 3 3
GA 2.5350 1.9065 1.8111
0.1582 1.3436
0.8510
Table 3.4: Optimized channel bandwidth for available
channels
Once the paths are selected then the NP hard problem of
optimal allocation of channels isleft,whichisveryimportant
to reduce the power consumption in the network. For this
we have used hybrid PSO-GSA optimization which is better
than genetic algorithm (GA) and GSA as GA is a local
optimization algorithm which sometime skips local minima
points during the search of minimum point whereas GSA is
the global optimization algorithm which checks every point
in search of global minima. The PSO-GSA combination uses
the strength of both local and global optimization algorithm
because PSO is also a local search algorithm.
The comparison of total transmitted power and path loss
component is shown in table 3.5. and results are verified for
different number of nodes and tabulated in 3.5.

Figure 3.6 path loss comparisons for different active nodes
This figure proved that with number of increase in active
nodes, path loss and transmitted power consumption also
increases but still with our proposed method i.e. PSO-GSO, it
is still very less than genetic algorithm.
4.Conclusion
The main challenge of IOT is the resource like energy,
memory, range constraint.Theselimitationaffecttherouting
in several ways. The best way of routing is multi hops.
Rather than sending the data direct to sink which consumes
large power, it must be sent through multiple nodes which
consumes less power for each node. The low processing
power and program memory require that the routing
process running on theIOTdevicesmust behighlyoptimized
and light-weight.
These constraint problems needs a solution for this non
linear problem which can be best described as an
optimisation problem. We proposed a hybrid optimisation
algorithm which is a combination of local and global search
optimisation. We combined them rather than cascading.
Table 3.5: output parameters comparison for different number of active nodes
Active
nodes

9 20 30 40
GA GSA PSO-
GSA
GA GSA PSO-
GSA
GA GSA PSO-
GSA
GA GSA PSO-
GSA
Path
Loss
Compo
nent in
dB
238.828
8 126.9
936
103.4
392
841.883
3
674.01
49
481.
8833
1.3580e
+03
1.053
2e+03
529.5
278
2.251
0e+0
3
1.601
3e+03
788.3
626
Total
Trans
mitted
Power
in mW
2.7049e
+41
2.708
7e+1
4
3.098
9e+13
1.4852e
+65
3.7573
e+35
1.54
66e+
25
1.2062e
+31
1.254
9e+67
1.254
9e+2
7
2.903
1e+6
4
2.508
1e+40
2.903
1e+2
4
Particle Swarm optimisation (PSO) and Gravitational Search
Algorithm (GSA) which are local and global optimisation
respectively are used tomakea hybridandeffectivesolution.
Previously Genetic Algorithm was used to allocate the
channels dynamically. In contrast to PSO-GSA, the genetic
algorithm is quite old optimization technique and proposed
optimisation algorithm is better than GA in terms of
convergence. So we have updated the work with this
algorithm and compared it with resultsofGA.Thetargetisto
minimize the power-consumption in IoT sensors used for
data transmission and connectivity. An improvement of
58.59% in path loss component is achieved by PSOGSA for
20 active nodes with same input data and total bandwidth
allocated.

REFERENCES
[1]. Quan Le; Thu Ngo-Quynh;Magedanz,T.,"RPL-based
multipath Routing Protocols for Internet of Things
on Wireless Sensor Networks," in Advanced
Technologies for Communications (ATC), 2014
International Conference on , vol., no., pp.424-429,
15-17 Oct. 2014
[2]. Sang-Hyun Park, Seungryong Cho, and Jung-Ryun
Lee, “Energy-Efficient Probabilistic Routing
Algorithm for InternetofThings,” Journal ofApplied
Mathematics, vol. 2014, Article ID 213106.
[3]. HyungWon Kim, "Low power routing and channel
allocation method of wireless video sensor
networks for Internet of Things (IoT)," in Internetof
Things (WF-IoT), 2014 IEEE World Forum on , vol.,
no., pp.446-451, 6-8 March 2014
[4]. Al-Turjman, F.; Gunay, M., "CAR Approach for the
Internet of Things," in Electrical and Computer
Engineering, Canadian Journal of , vol.39, no.1,
pp.11-18, winter 2016
[5]. Sung-Keun Lee , Jin-Gwang Koh and Chang-Ryul
Jung,” An Energy-Efficient QoS-aware Routing
Algorithm for Wireless Multimedia Sensor
Networks”, International Journal ofMultimedia and
Ubiquitous Engineering Vol.9, No.2 (2014), pp.245-
252
[6]. Chetana V Jyothi, Reema Sharma,” Reducing Power
and Interference Using a Routing and Multichannel
Allocation Approach in Wireless Sensor Networks
for Internet of Things (IoTs)”, International Journal
of Innovative Research in Computer and
Communication Engineering, Vol.3, Special Issue 5,
May 2015
[7]. Sastry Kompella, Member, IEEE, Shiwen Mao,
Member, IEEE, Y. Thomas Hou, Senior Member,
IEEE, and Hanif D. Sherali,” On Path Selection and
Rate Allocation for Video in Wireless Mesh
Networks”,
[8]. IEEE/ACMTRANSACTIONSON NETWORKING,VOL.
17, NO. 1, FEBRUARY 2009
[9]. Samia Allaoua Chelloug,” Energy-Efficient Content-
Based Routing in Internet of Things”, Journal of
Computer and Communications, 2015, 3, 9-20
[10]. Yong Ding, Li Xiao, "Video On-Demand
Streaming in Cognitive Wireless Mesh
Networks", IEEE TransactionsonMobileComputing,
vol.12, no. 3, pp. 412-423, March 2013
[11]. Roslin, S.E., "Genetic algorithm based
cluster head optimizationusingtopologycontrol for
hazardous environment using WSN," in Innovations
in Information, Embedded and Communication
Systems (ICIIECS), 2015InternationalConferenceon ,
vol., no., pp.1-7, 19-20 March 2015
[12]. A.S. Uma maheswari, Mrs. S. Pushpalatha,”
Cluster Head Selection Based On Genetic Algorithm
Using AHYMN Approaches in WSN”, International
Journal of Innovative Research in Science,
Engineering and Technology Volume 3, Special
Issue 3, March 2014
[13]. Kiranpreet Kaur1, Harjit Singh,” Cluster
Head Selection using Honey Bee Optimization in
Wireless Sensor Network” International Journal of
Advanced Research in Computer and
Communication Engineering Vol. 4, Issue 5, May
2015
[14]. R.Aiyshwariya Devi,M.Buvana,” Energy
Efficient Cluster Head Selection Scheme Based On
FMPDM for MANETs” International Journal of
Innovative Research in Science, Engineering and
Technology Volume 3, Special Issue 3, March 2014
[15]. Ebin Deni Raj,” An Efficient Cluster Head
Selection Algorithm for WirelessSensorNetworks –
Edrleach”, IOSR Journal of Computer Engineering
(IOSRJCE) ISSN: 2278-0661 Volume 2, Issue2(July-
Aug. 2012)
[16]. Nabil Ali Alrajeh, S. Khan, and Bilal Shams,”
Intrusion Detection Systems in Wireless Sensor
Networks: A Review” International Journal of
Distributed Sensor Networks Volume 2013.
[17]. Ajith Abraham, Crina Grosan, and Carlos
Martin-Vide,” Evolutionary Design of Intrusion
DetectionPrograms” International Journal of
Network Security, Vol.4, No.3, 2007
[18]. Ioannis Krontiris, Zinaida Benenson,
Thanassis Giannetsos, Felix C Freiling, Tassos
Dimitriou,” Cooperative Intrusion Detection in
Wireless Sensor Networks”, Wireless sensor
networks, Springer Berlin Heidelberg,2009
[19]. Djallel Eddine Boubiche and Azeddine
Bilami,” Cross Layer IntrusionDetectionSystemFor
Wireless Sensor Network” International Journal of
Network Security & Its Applications (IJNSA), Vol.4,
No.2, March 2012.
[20]. Shio Kumar Singh, M P Singh, and D K
Singh,” Intrusion Detection Based Security Solution
for Cluster-Based Wireless Sensor Networks”
International Journal of Advanced Science and
Technology Vol. 30, May, 2011
[21]. A.Anbumozhi, K.Muneeswaran,” Detection
of Intruders in Wireless Sensor Networks Using
Anomaly” IJIRSET Volume 3, Special Issue 3, March
2014.

PSO-GSA Tuned Dynamic Allocation in Wireless Video Sensor Networks for IOT

Recommended

More Related Content

What's hot (20)

Similar to PSO-GSA Tuned Dynamic Allocation in Wireless Video Sensor Networks for IOT (20)

More from IRJET Journal (20)

Recently uploaded (20)

PSO-GSA Tuned Dynamic Allocation in Wireless Video Sensor Networks for IOT