Jour of Adv Research in Dynamical & Control Systems, Vol.

11, 10-Special Issue, 2019

Cloud Assisted Internet of Things for Health

Care Monitoring based on Optimization
Techniques
Dr. Sam Abraham, Asst. Professor, Department of Computer Applications, Malankara Catholic College, Mariagiri,
Kanyakumari District, Tamil Nadu, India. E-mail: [email protected]
Dr.R.S. Anoop Sreekumar, Asst. Professor, Department of Computer Science, Malankara Catholic College, Mariagiri,
Kanyakumari District, Tamil Nadu, India. E-mail: [email protected]
Dr.S.U. Aswathy, Professor, Department of Computer Science and Engineering, Mangalam College of Engineering,
Ettumanoor, Kottayam, Kerala India. E-mail: [email protected]
Dr. Bibin Vincent, Professor, Department of Computer Science and Engineering, Mangalam College of Engineering,
Ettumanoor, Kottayam, Kerala India. E-mail: [email protected]
Abstract--- The Internet of Things (IoT) is an enormous internetwork of things. For creating the new service the
information’s are communicate through the wireless and cable connections. In today’s computing environment one
of the important modernism is the cloud computing Cloud computing and IoT plays a very inventive role in human
life. Monitoring the health parameter and maintain the health care of patients the Medical Technology is a widely
useful. In the previous work the stakeholder’s requirements, CPU utilization and execution time of the system are
concerned as the main issue. To overcome this proposed work contains various process such as: Stakeholder’s
devices, stakeholder’s request (tasks), cloud broker and network administrator. Here two optimization methods are
used in the research work such as, Bacteria Foraging Optimization (BFO) and Grey Wolf Optimization (GWO). An
experimental result shows that the details about the optimization techniques, execution time, data processing time
and system efficiency. In this novel method the total execution time rate is decreases and the system efficiency is
improved.
Keywords--- IoT, Cloud Computing, Stakeholders Devices, Big Data.

I. Introduction
Electronic devices, smart phones and tablets are the basic instrument of everyday life in the new era of
communication and technology. These are physically or wirelessly transmitted and this technology's development is
very big. Furthermore, the world's future generations are connected with the Internet of Things (IOT), which will
interface with phones, sensors, machines, cars and other "things" [1]."If we had PCs that knew everything there was
to know about things using information they'd assembled without any assistance from us, we'd have the option of
tracking and checking everything, and vastly reducing waste, losses, and costs. We'd know when things needed to be
replaced, fixed, or reviewed, and whether they were new or past their best. To do this, we need to engage PCs with
their own techniques. Victor [2] explains a data analysis method. It includes two cases where the first case focuses
mainly on analyzing the weather data and the second case concentrates the function of reducing the map. The map
decreases processing time and makes information visualization possible. Khan et al.[3] created some cloud
computing and Internet of Things (IOT) methods for healthcare applications such as intelligent hospitals, remote
medical facilities and medicine control. The primary objective of this technique is to analyze the state of the art gap
between the distinct concentrations of the mixed parts. One sort of heuristic algorithm is optimization forging
bacteria. Researchers have been drawing motivation from nature and ordinary animals for a significant length of
time to deal with complicated search problems of this current truth. Optimization is at the heart of many periodic
processes such as Darwinian growth, social insect group behavior, and other microbial creatures' foraging strategy
[4]. Because they are more likely to appreciate reproductive success, it is used to eliminate bad foraging techniques
and promote the spread of species characteristics with better foraging the natural selection algorithm. Since a
foraging organism or creature makes a significant move to expand the consumption of vitality per unit of time spent
on foraging, taking into account, for example, every requirement introduced by its own physiology, detection and
intellectual capacities, the common search methodology can prompt progress and this idea can basically be used to
solve real-world optimization problems [5]. Passino anticipated an optimization procedure called the Bacterial
Foraging Optimization Algorithm (BFOA) in accordance with this idea.

DOI: 10.5373/JARDCS/V11SP10/20192889
ISSN 1943-023X 931
Received: 23 Aug 2019/Accepted: 20 Sep 2019
Jour of Adv Research in Dynamical & Control Systems, Vol. 11, 10-Special Issue, 2019

This approach resembles a partial model of random walk [6]. Inspired by the Gray Wolf's level of management
and hunting schemes, Mirjalili suggested Gray Wolf Optimization Algorithm, creating the social progression and
predatory conduct of the gray wolf [7]. GWO has a fundamental framework and fast mixture compared to Immune
Particle Swarm Optimization (IPSO), Ant Colony Algorithm (ACA), GA, and Biogeography-based Optimization
(BBO). Recently, GWO has been used to improve the presentation of Convolution Neural Network (CNN) models
for classification. Kumaran N et al.[8] suggest a mixed CNN-GWO strategy for the recognition of human operations
from unconstrained video recording. Trial approval of this technique demonstrates better viable results with 99.9
percent accuracy in recognition of human operations. Mustaffa Z et al.[9] also use GWO to prepare Least Squares
Support Vector Machine (LSSVM) to determine the value and to present a model for hybrid prediction. Section 2
explains the suggested model of the scheme, section 3&4 explains the optimization of Bacteria Foraging and Gray
Wolf, chapter 5 describes the results and evaluation, followed by a conclusion in chapter 6.

II. Methodology
The overall procedure of the planned architecture is shown below. This architecture consist of four stages they
are stakeholders, IHS devices, IHS tasks, cloud broker and network administrator. The communication was done
between HIS and cloud which is mainly for executing the network communication. The architecture is shown in
fig 1.
PC (Personal Computer), Laptop, Smartphone, Tablet, is the various devices used by IHS. The patients
requirements are send to cloud environment. Initially the patient detail was collected and further provides the
various medical facilities such as diseases diagnosis and electronic medical records (EMR). The transfer and getting
tasks from the cloud service are done by cloud broker. Every network has many application servers and providing
the SaaS (Service as a Service). The tasks in each stakeholder can be allocated and each application server has a
resource collection that can be allocated for the tasks to come. All networks have the network administrator that
will be responsible for transforming information between the networks in the cloud between the other networks and
the classified servers. The network administrator's primary aim is to apply optimization models. (Bacteria Foraging
Optimization & Grey Wolf Optimization) This will help to get VMs in the cloud to develop a task scheduling
process that is direct to the total IHS request time and maximum use of resources.

Figure 1: Architecture of the Proposed Method

DOI: 10.5373/JARDCS/V11SP10/20192889
ISSN 1943-023X 932
Received: 23 Aug 2019/Accepted: 20 Sep 2019
Jour of Adv Research in Dynamical & Control Systems, Vol. 11, 10-Special Issue, 2019

III. Bacterial Foraging Optimization

Bacterial foraging optimization algorithm (BFOA) is commonly recognized global optimization. The social
foraging behavior of Escherichia coli stimulates BFOA.
The locomotion is achieved through the use of pair of tensile flagella in the foraging of the real bacteria. The
basic operations performed by bacterium flagella at the time of foraging the tumble or swimming are helpful for an
E. coli bacterium to tumble or swim. When each flagellum pushes the E.coli rotates the flagellum in the direction of
the clock. When the flagella are in block-clockwise direction, the bacterium swims at a rapid rate. The end result is
the autonomous movement of the flagella and the bacterium tumbles with less tumbling. It commonly tumbles to
discover a nutrient gradient in a harmful location. The bacteria experience chemotaxis in which they like to move
towards a gradient of nutrients and prevent harmful environments. Normally the organism will move the bacteria for
a longer range in a friendly virtual environment. The clockwise and counter-clockwise movement of a bacterium
occurs in a nutrient solution and also provides an account of most of BFOA's significant applications to date.
The bacteria breaking in the center to an exact reproduce themselves in the presence of the appropriate
temperature. The bacterium increased in length and if the food is insufficient. This event encouraged Passino to
identify a reproductive event in BFOA. Chemotactic progress may be destroyed as a result of sudden environmental
change or attack, and a group of bacteria may move to some other location. Where all the bacteria are killed in a
region or a group is dispersed into a new part of the environment, this is the elimination-dispersal in the real
bacterial population. The BFOA's four significant steps are.
a. Chemotaxis
This process simulates the movement of an E.coli cell based on the swimming and tumbling of flagella. This is
going to move biologically in two different ways. It can swim or tumble in the same direction for a certain period of
time and alternate over the entire lifetime between these two operating modes. The computing chemotax can be
depicted by the motion of the bacterium

(1)
Where

indicates A random-directed vector with components in [ -1, 1 ].

( j, k, l) represents i bacterium in chemotactic, k-th reproductive and l-th dispersal steps.

C(i) is the size of the move in the burst-specified random direction.
b. Swarming
For different mobile bacteria species like E.coli and Typhimurium observed group behavior that was interesting.
In semisolid nutrient medium where complex and stable spatio-temporal patterns are formed (swarms). When put in
a semisolid matrix with a single nutrient chemo-effector, a group of E.coli cells assemble in a traveling ring by
going up the nutrient gradient. The cells release an attractive a spertate when stimulated by a elevated concentration
of succinate, which enables them to aggregate into organizations and move as concentrated models of elevated
bacterial density swarms. The signaling of cell to cell in E. The following function may represent E.coli swarm.

(3)
Where,

, P( j, k, l)) is the actual objective function,

S is the total number of bacteria, p is the number of variables to be optimized, and is p-

dimensional search domain. are various coefficients which should
be selected correctly.

DOI: 10.5373/JARDCS/V11SP10/20192889
ISSN 1943-023X 933
Received: 23 Aug 2019/Accepted: 20 Sep 2019
Jour of Adv Research in Dynamical & Control Systems, Vol. 11, 10-Special Issue, 2019

c. Reproduction
The healthiest bacteria eventually die when each healthier bacterium divides into two bacteria, which are then
placed in the same place. This keeps the swarm size stable.
d. Elimination and Dispersal
Where there are gradual or sudden changes in the local setting, bacterial population lives may happen due
to multiple factors. All the bacteria in a region are killed or a group is dispersed in such a manner into a new location
that these events can occur. Some bacteria are exchanged indiscriminately with an extremely low probability to
simulate this occurrence in BFOA while the new alternates are initialized arbitrarily over the search space.
Algorithm Steps
Parameters:

 Step 1: p, S, , Ns, Nre, Ned, Ped, C(i)(i=1,2…S) , are initialized

 Step 2:Loop of elimination-dispersal: i=i+1
 Step 3: Loop of reproduction: k=k+1
 Step 4: loop of Chemotaxis:j=j+1:
a.To obtain chemotactic step take i=1,2…s
b. Fitness value calculated J (i, j, k, l)

Let, = )+
c.Marks the value Jlast=J (i, j, k, l)
d. Tumble: with each element ( i)= 1,2,….,p, a random number on [-1,1].
e.shift: Let

(4)
This is a result of C(i).
a. calculate J (i, j +1, k, l) and let
= )+ (5)
b. Swim
1. Let m=0.
2. While m< Ns.
 Let m=m+1.
 If J (i, j +1, k, l) < Jlast,
 let Jlast = J (i, j +1, k, l) and

let (6)
 Else, Let m=Ns. This is the ending of the while statement.
a. Go to next bacterium (i+1) if i S.
 Step 5: If j<Nc go to step 4.

(7)
 Step 6: If re k < Nre , go to step 3.
 Step 7: Elimination-dispersal: for j = 1,2 ..., S with probability Ped, each bacterium eliminates and
disperses. If ed l < Ned, go to step 2; if not, end.

IV. Gray Wolf Optimization

Grey wolf optimization is a swarm shrewd procedure created by Mirjalili etal., 2014, which imitates the wolves '
leadership hierarchy for chasing their group.

DOI: 10.5373/JARDCS/V11SP10/20192889
ISSN 1943-023X 934
Received: 23 Aug 2019/Accepted: 20 Sep 2019
Jour of Adv Research in Dynamical & Control Systems, Vol. 11, 10-Special Issue, 2019

Grey wolf has a place to live in a pack with the family of Canidae. They have a predominantly exacting social
progression; the leader is a male or female, Alpha (α). Generally, the alpha is responsible for basic leadership. The
pack should trail the sets of the prevailing wolf.
The Betas (β) are subordinate wolves in fundamental management that assist the alpha. The beta is an alpha
guide and pack discipliner. The gray wolf in the reduced position is Omega (ω) that must present all other
predominant wolves. If a wolf is unlikely to be an alpha or beta or omega, it is called delta. Delta wolves overwhelm
omega and report alpha and beta to wolves. The methods of chasing and the wolves ' social hierarchy of importance
is scientifically proven to create GWO and optimize. The calculation of the GWO is tested with conventional test
works showing that it has prevailing characteristics of inquiry and exploitation than other swarm intelligence
technologies. In addition, the GWO was efficiently linked to take care of various problems of engineering
optimization. Moreover, the greater part of the swarm intelligent systems used to take care of the issues of
improvement cannot have the leader organizing over the entire time frame. In GWO, where the gray wolves have a
natural management structure, this downside is rectified. Additionally, this calculation has a few parameters that are
just and easy to update, making it unrivaled than previous ones. Due to the GWO calculation's flexible properties,
attempts were made to execute GWO in order to solve the optimization problems.
Algorithmic Steps
 Step 1: the search agents (Gs), design variable size (Gd), vectors a, A, C and maximum number of iteration
(itermax).are the GWO parameters which are initialized.

(8)

(9)
The rates of are reduced from 2 to 0.
 Step 2: based on size of the pack the wolves generated arbitrarily. Mathematically , this is represented as,

(10)
Where,

is the original value of the jth pack of the ith wolves.

 Step 3: Estimate the fitness value of each hunt agent

(11)

 Step 4: recognize the best hunt agent ( ), the second best hunt agent ( ) and the third best agent ( )

(13)

(14)

(18)
 Step 5: Update the location of the current hunt agent.

(19)

DOI: 10.5373/JARDCS/V11SP10/20192889
ISSN 1943-023X 935
Received: 23 Aug 2019/Accepted: 20 Sep 2019
Jour of Adv Research in Dynamical & Control Systems, Vol. 11, 10-Special Issue, 2019

 Step 6: approximate the fitness value of every hunts.

 Step 7: renew the value

 Step 8: Check for stopping condition Iter=Itermax, if yes print best value otherwise loop to step 5.

V. Experimental Analysis
In this work two different algorithms such as: Bacteria Foraging Optimization and Grey Wolf Optimization are
used for evaluating the performance of the proposed work. Cloud computing environment used a single data center,
a broker and user, these work are implemented by MATLAB.

Figure 2: Experimental Results of Both Algorithms

Using the FCFS algorithm the allotment of VMs(Virtual Machine) to the hosts is done, based on the machine
shared policy allocating the task to the virtual machine is performed., consequently the task are executed serially in
each VM. For that reason number of incoming tasks did not affect execution time by using the space shared policy.
In MATLAB environment, evaluate experiments were performed in two cases using single VM and double
VMs. Here calculates the performance metrics such as, CPU utilization, Turn-around time and Waiting time and
also compare to the Bacteria Foraging Optimization and Grey Wolf Optimization.

VI. Conclusion
The cloud computing and IoT is considered as the emerging technology in today’s environment. This proposed
method mainly focused on stakeholder’s requirements, CPU utilization and execution time. With the help of BF and
GWO algorithm the execution time was reduced and thereby the performance of the system was improved. The
obtained results are compared with the BF and GWO algorithm the GWO algorithm provides better results. Here for
processing two simulations are used, such as MATLAB and Cloud Sim was done. These two platforms provide best
performance analysis of the system. Total execution time rate is 28.5% compare to the existing system and also
improve the system efficiency.

References
[1] Azzawi, Mustafa Abdullah, Rosilah Hassan, and Khairul Azmi Abu Bakar. "A review on Internet of Things
(IoT) in healthcare." International Journal of Applied Engineering Research 11, no. 20 (2016): 10216-
10221.
[2] Victor Chang, “Towards data analysis for weather cloud computing”, Knowledge Based Systems journal,
Vol.127, pp.29-45, July 2017.

DOI: 10.5373/JARDCS/V11SP10/20192889
ISSN 1943-023X 936
Received: 23 Aug 2019/Accepted: 20 Sep 2019
Jour of Adv Research in Dynamical & Control Systems, Vol. 11, 10-Special Issue, 2019

[3] Ashraf Darwish, Aboul Ella Hassanien, Mohamed Elhoseny,Arun Kumar Sangaiah,KhanMuhammad, “The
impact of the hybrid platform of internet of things and cloud computing on healthcare systems:
opportunities, challenges, and open problems”, Journal of Ambient Intelligence and Humanized
Computing, pp.1-16,December 2017.
[4] Kim, Dong Hwa, Ajith Abraham, and Jae Hoon Cho. "A hybrid genetic algorithm and bacterial foraging
approach for global optimization." Information Sciences 177, no. 18 (2007): 3918-3937.
[5] Das, Swagatam, Arijit Biswas, Sambarta Dasgupta, and Ajith Abraham. "Bacterial foraging optimization
algorithm: theoretical foundations, analysis, and applications." In Foundations of Computational
Intelligence Volume 3, pp. 23-55. Springer, Berlin, Heidelberg, 2009.
[6] Dasgupta, Sambarta, Swagatam Das, Ajith Abraham, and Arijit Biswas. "Adaptive computational
chemotaxis in bacterial foraging optimization: an analysis." IEEE Transactions on Evolutionary
Computation 13, no. 4 (2009): 919-941.
[7] Ren, Yixuan, Tao Ye, Mengxing Huang, and Siling Feng. "Gray Wolf Optimization Algorithm for Multi-
Constraints Second-Order Stochastic Dominance Portfolio Optimization." Algorithms 11, no. 5 (2018): 72.
[8] Kumaran, N.; Vadivel, A.; Kumar, S.S. Recognition of human actions using CNN-GWO: A novel
modeling of CNN for enhancement of classification performance. Mutimed. Tools Appl. 2018, 1, 1–33.
[9] Mustaffa, Z.; Sulaiman, M.H.; Kahar, M.N.M. Training LSSVM with GWO for price forecasting. Electron.
Vis. 2015, 1–6.
[10] S. Ghosh, P. K. Panigrahi, and D. R. Parhi, “Analysis of FPA and BA meta-heuristic controllers for optimal
path planning of mobile robot in cluttered environment,” IET Sci. Meas. Technol., vol. 11, no. 7, pp. 817–
828, 2017.
[11] Mohamed .E, Alaa. T, Ahmed .F, and Aboul Ella. H, “K-Coverage Model based on Genetic Algorithm to
extend WSN lifetime”, IEEE Sensors Letters, Vol.1, No.4, pp.1-4, August 2017.
[12] Mofijul. I, J. Islam and A. Razzaque, “A Genetic Algorithm for Virtual Machine Migration in
Heterogeneous Mobile Cloud Computing”, International Conference on Networking Systems and Security,
February 2016.
[13] Nagy Ramadan Darwish, Ahmed A. Mohamed and Bassem S. M. Zohdy, “Applying Swarm Optimization
Techniques to Calculate Execution Time for Software Modules”, International Journal of Advanced
Research in Artificial Intelligence, Vol. 5, No.3,pp.12-17, 2016.
[14] Ahmed .A, Mohamed .E, Ahmed S. Salama and A.M. Riad, “A Machine Learning Model for Improving
Healthcare services on Cloud Computing Environment”, Elsevier Publications,Vol.119,pp.117-128,April
2018.
[15] Seyed. E Dashti and Amir Masoud. R, “Dynamic VMs placement for energy efficiency by PSO in cloud
computing”, Journal of Experimental & Theoretical Artificial Intelligence, Vol.28, pp.97-112, April2015.
[16] Ranjan Kumar, G. Tharwat Sahoo, “Cloud Computing Simulation Using CloudSim”, International Journal
of Engineering Trends and Technology (IJETT), Vol. 8 ,No. 2,pp.82-86,February 2014.
[17] Long, Wen, Jianjun Jiao, Ximing Liang, and Mingzhu Tang. "Inspired grey wolf optimizer for solving
large-scale function optimization problems." Applied Mathematical Modelling, 60 (2018): 112-126.

DOI: 10.5373/JARDCS/V11SP10/20192889
ISSN 1943-023X 937
Received: 23 Aug 2019/Accepted: 20 Sep 2019