You are on page 1of 10

Is the cloud computing real or hype?

Affirmation momentous traits


of this technology by proffering maiden scenarios

Meysam Panahi
1
, Mehdi Darbandi
2
, Mohammad Abedi
2
, Ali Hamzenejad
3
, Mohsen
Kariman Khorasani
4
1
Department of Management System and Productivity, Faculty of Industrial Engineering, Tehran South Branch,
Islamic Azad University (IAU), Tehran, Iran
2
Department of Electrical Engineering and Computer Science at Iran University of Science and Technology
(IUST), Tehran, Iran
3
Department of Management System and Productivity, Faculty of Industrial Engineering, Tehran South Branch,
Islamic Azad University (IAU), Tehran, Iran
4
Department of Communication Engineering, Islamic Azad University, Gonabad Branch, Gonabad, Iran

Abstract Cloud computing is a massively central
advancement in the technique that businesses and users devour
and work on computing. It's a elementary modify to an
prepared model in which applications don't subsist out their
lives on a specific section of hardware and in which possessions
are more supplely deployed than was the historical standard.
It's a primary shift to expansion and utilization model that
replaces hard-wired, proprietary associations surrounded by
software components and the clients of those components with
unimportant Web services and Web-based software
admittance. As difficulty under the virtualization is how to
allocate resources to every application on-demand and in
response to time-varying workloads. As cloud service can be a
public, private or mixture of both phenomena, cloud
computing is also functioning as platform as service, that
allow users to implement as virtual system for application
development. Among multiple platforms' choices, a new user
gets confused about the picking of platform for his own
requirement as per cost and performance trade off. Although
the cloud computing architecture is so elastic that new
numbers of data and processing recourse can easily added but
in cloud computing architecture, the massive date for
processing will situated in some unknown location and used is
constrained to access that data according to provider
agreements. Same way a difficulty in the virtualization is
allocation of requires resources to each request according to
on-demand manner handling with efficient jobs fulfillment
with in time. With analyses to date is, that the architectures of
cloud computing do not reflect the user perspectives and with
the technical requirements fully but it has potential to capture
the market of development and dynamic scaling to datacenters.
Cloud computing is making it possible to separate the process
of building an infrastructure for service provisioning from the
business of providing end user services. Today, such
infrastructures are normally provided in large data centers
and the applications are executed remotely from the users. One
reason for this is that cloud computing requires a reasonably
stable infrastructure and networking environment, largely due
to management reasons. Networking of Information (NetInf) is
an information centric networking paradigm that can support
cloud computing by providing new possibilities for network
transport and storage. It offers direct access to information
objects through a simple API, independent of their location in
the network. This abstraction can hide much of the complexity
of storage and network transport systems that cloud
computing today has to deal with. Cloud Computing is
becoming a well-known buzzword nowadays. Many companies,
such as Amazon, Google, Microsoft and so on, accelerate their
paces in developing Cloud Computing systems and enhancing
their services to provide for a larger amount of users.
However, security and privacy issues present a strong barrier
for users to adapt into Cloud computing systems. In this
paper, at first we consider significant influences of this
technology on some of the biggest companies and
organizations all over the world, after that we present
our new Kalman filter for estimation and prediction of
permeating hackers. This filter is a new two-stage
Kalman filter that is optimized and became more
efficient for this purpose. With the use of this paper we
can estimate the behaviors of hackers and/or estimate
the crowd sections of cloud, so that we can avoid
crashing of the network or hanging of such networks.
Keywords- Virtual or reality, Cloud scenarios, Virtual
machine.

I. INTRODUCTION:
Cloud computing is a general term for anything that
involves delivering hosted services over the internet. A
cloud service has three distinct characteristics that
differentiate it from traditional hosting. It is sold on
demand, typically by the minute or the hour; it is elastic
-- a user can have as much or as little of a service as
they want at any given time; and the service is fully
managed by the provider (the consumer needs nothing
but a personal computer and internet access). It is one
of those emerging technologies that everyone knows
they want to get into, but no one seems to know why or
what the real benefit is. The factor that makes the
current trend more powerful is: cost-cutting. Cloud in
this case, is really synonymous with "the internet."
Cloud computing is not only a technology; it is the
conversion of a set of technologies to be offered in the
form of an internet- or network-based service rather
than as asset purchases. In general, cloud computing
customers do not own the physical infrastructure,
instead avoiding capital expenditure by renting usage
from a third-party provider. They consume resources as
a service and pay only for resources that they use. A
side-effect of this approach is that overall computer
usage rises dramatically, as customers do not have to
engineer for peak load limits. In addition, increased
high-speed bandwidth makes it possible to receive the
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 116
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
same response times from centralized infrastructure at
other sites.
Cloud computing delivers computing as a utility to
users worldwide, a consequence of this model is that
cloud data centers have high deployment and
operational costs, as well as significant carbon
footprints for the environment. We need to develop
Green Cloud Computing (GCC) solutions that reduce
these deployment and operational costs and thus save
energy and reduce adverse environmental impacts. In
order to achieve this objective, a thorough
understanding of the energy consumption patterns in
complex Cloud environments is needed. Cloud
computing presents a new model for IT service
delivery and it typically involves over-a-network, on-
demand, self-service access, which is dynamically
scalable and elastic, utilizing pools of often virtualized
resources. Through these features, cloud computing has
the potential to improve the way businesses and IT
operate by offering fast start-up, flexibility, scalability
and cost efficiency. Even though cloud computing
provides compelling benefits and cost-effective options
for IT hosting and expansion, new risks and
opportunities for security exploits are introduced.
Standards, policies and controls are therefore of the
essence to assist management in protecting and
safeguarding systems and data. Management should
understand and analyze cloud computing risks in order
to protect systems and data from security exploits.
ICT systems have been investigated for flexible
systems configuration, systems operation cost
reduction, environmental impact reduction, etc. Cloud
computing has attracted attention as technology that
solves these. In the U.S., business Cloud services, such
as Amazon EC2/S3, Google Apps, Force.com, and
Windows Azure, are gaining more and more users.
Additionally, study of Cloud computing, such as a
governmental I-Japan strategy and a start of the smart
Cloud study group of the Ministry of Internal Affairs
and Communications, is progressing rapidly in Japan.
However, the investigation on the present Cloud
computing is mainly focused on the service side, while
the security side has not been sufficiently looked at.
The security perception by the social viewpoints of a
user's vague uneasiness has especially been
insufficiently investigated.
Trust is one of the main obstacles to widespread Cloud
adoption. In order to increase trust in Cloud computing,
we need to increase transparency and accountability of
data in the Cloud for both enterprises and end-users.
However, current system tools are unable to log file
accesses and transfers effectively within a Cloud
environment.
As Cloud computing-based technology has grown, so
have the number of services offered over the Cloud.
Due to this rapid growth in Cloud-based technology,
service providers must be able to deploy Cloud service
environments very quickly and very easily. To help
resolving this issue, an Ezilla, deploy private Cloud
toolkit, has been developed by the Pervasive
Computing Team at the National Center for High-
Performance Computing (NCHC). This effort of Ezilla
integrates the de facto Cloud middleware, Web-based
Operating System (WebOS), and coordinated Cloud
infrastructure services (storage, computing, and
networking) to form a virtual computer in distributed
computing environment. With a click, Cloud users can
easily customize and configure the specified virtual
clusters via Ezilla toolkit. The main feature of Ezilla is
simplifying a lot complexity of utilizing Clouds.
Cloud computing is a computing platform with the
backbone of internet to store, access the data and
application which is in the cloud, not in the computer.
The biggest issue which should be addressed in cloud
computing are security and privacy. Outsourcing data
to other companies worries internet clients to think
about the privacy data. Most Enterprise executives
hesitate to use cloud computing system due to their
sensitive enterprise information.
Cloud computing, as an emerging computing model, is
attracting more and more attention of industry and
academia. Though it is of significant importance to
carry out studies on such programming models, so far,
there are few programming models for cloud
computing. Because various kinds of cloud platforms
are developed independently, it is necessary to have the
support of cloud platform in making cloud
programming model be able to deploy application to
the cloud. These are just part of the great performance
of new technology, known as Cloud computing that is
named also as "the next big thing" [1-9].
II. CONSIDERING HIGH IMPACTS OF
CLOUD COMPUTING ON DIFFERENT
INDUSTRIES ASE OF USE
In two past sections of the paper, we define some of
the basic and fundamental principles of cloud and also
we tell about some of its advantageous. Now we want
imply into, the major applications of this technology.
After that when we understand the importance of this
technology, we talk about some techniques and
algorithms which can be uses for improving the
security aspect of such network; for example, we can
used Kalman Filter for prediction and estimation the
amount of users that can be allowed to logging into
special organization account. At the end of this paper
we suggest and demonstrate new Kalman estimator
which is more efficient and more reliable for such
these networks.
The introduction of data networks within an aircraft
has created several service opportunities for the air
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 117
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
carriers. Using the available Internet connectivity, the
carriers could offer services like Video-on-Demand
(VoD), Voice-over-IP (VoIP), and gaming-on-demand
within the aircraft. One of the major road blocks in
implementing any of these services is the additional
hardware and software requirements. Each service
requires dedicated hardware resources to run
appropriate software components. It is not possible to
accommodate every hardware component within the
aircraft due to space, power, and ventilation
restrictions. Also, it is economically not viable to
install and maintain hardware components for every
aircraft. One solution is to use cloud computing. Cloud
computing is a recent innovation that is helping the
computing industry in distributed computing. Cloud
computing allows the organizations to consolidate
several hardware resources into one physical device.
The Cloud computing concept helps organizations in
reducing the overall power consumption and
maintenance costs. The cloud computing concept
could be extended to the Aircraft Data Network
environment with every aircraft subscribing to the
cloud resources to run their non mission-critical
applications.
The National Institute of Standards and Technology
(NSIT) and the cloud Security Alliance define cloud
computing as a Model for enabling convenient, on-
demand network access to a shared pool of
configurable computing resources (e.g., networks,
servers, storage, applications and services) that can be
rapidly provisioned and released with minimal
management effort or service provider interaction.
Gartner defines cloud computing as: A style of
computing where massively scalable, IT-enabled
capabilities are provided as a service across the
Internet to multiple external customers. Cloud
computing refers to both the applications delivered as
services over the Internet and the hardware and
systems software in the data centres that provide those
services. It facilitates sharing of technological
resources, software and digital information across
multiple platforms, clients and users on Internet.
The cloud computing system can be conceptually
divided into two sections: the front end and the back
end, which connect to each other through a network,
usually the Internet. The front end is the interface by
the user for interaction with the back end. The front
end or user interface could be any device such as a
desktop, laptop, computer terminal, mobile devices,
smart phone, PDA, etc. and this could be located
anywhere. The back end is the "cloud" section of the
system which is the collection of servers connected on
a public/private network.
The front end includes the client's computer or
network and the application required to access the
cloud computing system. It is not necessary that all
cloud computing systems have the same user interface.
Services like web-based e-mail programmes use
existing web browsers like Google Chrome, Internet
Explorer, Firefox, Safari, etc. At the back end are
various computers, servers and data storage systems
that create the "cloud" of computing services.
Generally, each application is required to have its own
dedicated server. A central server administers the
system, monitoring traffic and user demands to ensure
that all applications and the network access are
running as required. The system follows a set of rules
called protocols and uses a special kind of software
called middleware which allows the networked
computers to communicate with each other.
Currently, there are three dominant cloud computing
service models: Infrastructure as a Service (IaaS),
Software as a Service (SaaS) and Platform as a Service
(PaaS). The service models are dynamically changing
as cloud providers come out with new offerings
focussed on being competitive, increase market share,
each with the aim to becoming one-stop shop as
feasible. The service models are explained below [1, 5,
15, and 16]:



Fig. 1: People all around the world can use Cloud
Computing and its software and hardware resources for their
own purposes or sharing information with anyone that they
want.

Infrastructure as a Service (Iaas): This model of
services typically includes the core IT Infrastructure
Services such as operating systems, data storage, web
servers, and edge caching services. This model
provides the capability to provision processing,
storage, networks and other fundamental computing
resources by offering the customer the ability to
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 118
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
deploy and run arbitrary software, which can include
operating systems and applications. Platform as a
Service (PaaS): This model provides the application
building blocks which may include: workflow,
document management, data services, APls, fabric,
proprietary development languages. This model
provides the capability to deploy onto the cloud
infrastructure customer-created or acquired
applications created using programming languages and
tools supported by the provider. PaaS enables
independent software vendors (ISVs) to develop,
deploy, and manage applications without incurring
upfront cost for buying the platform hardware or
software. PaaS platform has evolved from software as
a service (SaaS) and infrastructure as a service (IaaS).
Software as a Service (SaaS): This model provides
complete applications which are sold on a subscription
model for a specific period: Examples of software
provided through SaaS model are CRM, ERP, e-mail,
Calendar, Internet File Stores, Spam filters, etc. This
model provides the capability to use the provider's
applications running on cloud infrastructure. The
applications are accessible from various client devices
through a thin client interface such as a web browser
(e.g., web-based e-mail). SaaS saves customers the
cost of buying licenses and running programs on their
own computers.
Is cloud computing real or hype? The answer is: it is
somewhere in between. We have to go beyond the
hype created by the vendors and look at the reality of
how it can add value to the enterprise while mitigating
the relevant risks. As per Gartner research, every new
IT usage moves through different phases from
conception to obsolescence. Thus, the peak of inflated
expectations leads to the trough of disillusionment
which is followed by the slope of enlightenment and
the plateau of productivity. Currently, cloud
computing is in the peak phase so you can expect to
find lot of hype. However, it is expected to transition
into the enlightenment or growth phase shortly. A
sample of findings from latest surveys and study made
by consultancy and market research firms highlights
the importance and projected growth of cloud
computing.

Global Scenario:
1.Cloud services revenue to touch $149 billion in
2014. $55 billion forecasted worldwide revenue from
public IT cloud services alone.
2.Cloud services cost less than traditional outsourced
services, with savings ranging from 20% to 50%
depending on the type of service offered.
3.30% is the rate at which cloud computing will grow
in 2011, or more than five times the rate of IT industry
as a whole.
4.2.3 million (Net) new jobs will be created by cloud
on a cumulative basis from 2010-2015.
5.The impact of cloud computing will be very high on
the nearly $60 billion outsourcing sector, whose
mantra is cost savings. This sector has little choice but
to include cloud computing as part of their service
portfolio.

Indian Scenario:
1.India is ahead of US in cloud adoption. Top cloud
users today are Brazil (27%), Germany (27%), India
(26%), US (23%).
2.Cloud computing market in India is expected to
cross $1.08 billion by 2015, from $110 million in
2010.
3.Of the projected $4.5 billion total cloud computing
market in India by 2015, private cloud will account for
$3.5 billion.
4.Cloud computing will generate about 100,1XXI
additional jobs and save about 50% of cost of IT
operations for Indian enterprises.
5.The CEO and MD of India's second largest software
exporter says: "India needs a policy framework for the
new service that enables companies to share IT
infrastructure and cut costs. Efforts are on for such a
policy and the Confederation of Indian Industry (CII)
will submit a draft paper to the government. The
opportunities are huge for India for providing citizen
services and the right regulatory frame work will
accelerate the growth. Some of the issues like data
privacy and security should be addressed properly,
which is possible only with a regulatory framework.
Regulatory framework would give confidence that the
service providers will provide the service securely and
reliably. Our company has already appointed 2,000
experts to work on this new service".
6.India's No. 3 outsourcing firm looks at cloud
computing as a "game changer". It is building data
centers in India and is implementing private clouds in
partnership with other IT firms.
7.The cloud has the potential to transform business
ecosystems that are relatively under-penetrated by IT
due to high capital requirements, such as government,
healthcare and education.
BUSINESS Analyst deal with data/information in
myriad forms for assurance reviews, analysis and
decision-making. The location of digital data can be
traced to computers and servers either at identified
offices of clients or vendors. However, of late, there is
increasing talk of data being stored on the cloud and
usage of cloud computing which leads to new areas of
concern from perspective of risk, security and controls.
Enterprises are increasingly using cloud computing
due to business benefits it offers. As with any
deployment of IT by enterprises, cloud computing
offers both challenges and opportunities to business
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 119
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
analyst. The challenges are in terms of the inherent
risks of accessing data on the cloud and the impact on
the way services are provided. The opportunities are
the way in which cloud itself can be used to deliver
services of assurance/consulting to clients. It is
imperative that business analyst understand concepts
of cloud computing, service offerings, deployment
models and the related risks and controls for two
purposes:
1. Impact on services provided to clients by
accessing relevant data at clients offices or
remotely;
2. Impact on how services are/will be provided
within their own offices using cloud
computing.

Analysis of findings above highlight that cloud
computing is expected to be the next big wave in IT.
This clearly establishes that cloud computing is
emerging as the next big IT service for its pay-as-you-
go model, which will eliminate capital intensive
investment especially for the Small and Medium
Enterprises by minimizing investment in IT
infrastructure. There is growing recognition of the
power of cloud computing by many CIOs who are
looking to implement it in their own IT departments.
However, with all the noise in the market about cloud
computing, it is critical to remove the hype from the
reality, identify the right strategy and know where to
begin. There is no running away from cloud
computing as it is increasingly becoming all pervasive.
Business Analyst needs to understand the services
which can be delivered using the power of cloud
computing so as to be empowered to explore how to
convert challenges of the cloud to opportunities for
themselves and their clients.
Cloud computing and related cloud services generate
new ways of thinking about IT computing architecture
and service delivery models. With cloud, everything
becomes a service thus enabling enterprises to create
new initiatives without a huge IT investment. It
transforms capital expenditure to operating expenses
payable based on use. Cloud computing offers new
and unique business benefits which are expected to
change the way businesses collaborate, compete,
operate and deliver services. Hence, the number of
enterprises who use cloud computing will keep
increasing in the near future.
Finally, writers of this paper predict in near future lots
of people involved in this technology. Engineers and
designers must play their role more accurately to get
the interest of more users and provide them their needs
[1 to 5 - 16].

III. NEW KALMAN STATE ESTIMATOR
Now, we want to suggest and demonstrate our
new Kalman filter for the purpose of improving
the security and reliability of such networks. The
problem of interest is described by the discretized
equation set [17]:
x
k k k k k k
W U B X A X + + =
+1
(1)
u
k k k k
W U C U + =
+1
(2)
k k k k
V X H Z + = (3)
Where
n
k
R X e is the system state,
m
k
R U e and
p
k
R Z e are the input and the measurement
vectors, respectively. Matrices
k
A ,
k
B ,
k
C and
k
H are assumed to be known functions of the
time interval k and are of appropriate dimensions.
Matrix
k
C is assumed nonsingular. The process
noises
x
k
W ,
u
k
W and the measurement noise
k
V
are zero-mean white Gaussian sequences with the
following covariances:
kl
x
k
' x
l
x
k
Q ] ) W ( W [ E o = ,
kl
u
k
' u
l
u
k
Q ] ) W ( W [ E o = ,
kl
xu
k
' u
l
x
k
Q ] ) W ( W [ E o = ,
kl k
'
l k
R ] V V [ E o = , 0 = ] V W [ E
'
l
x
k
and
0 = ] V W [ E
'
l
u
k
, where
'
denotes transpose and
kl
o
denotes the Kronecker delta function. The initial
states
0
X and
0
U are assumed to be uncorrelated
with the sequences
x
k
W ,
u
k
W and
k
V . The initial
conditions are assumed to be Gaussian random
variables with
0 0
X

] X [ E = ,
x '
P ] X X [ E
0 0 0
= ,
0 0
U

] U [ E = ,
u '
P ] U U [ E
0 0 0
= ,
xu '
P ] U X [ E
0 0 0
= .
As we can see in (3) the input vector
k
U is not
observable through the measurement process.
Treating
k
X and
k
U as the augmented system
state, the AUSKE is described by
) X H Z ( K X X
Aug
k | k
Aug
k k
Aug
k
Aug
k | k
Aug
k | k 1 1 1 1 1 1 1 + + + + + + +
+ =
(4)
Aug
k | k
Aug
k
Aug
k | k
X A X =
+1
(5)
1 '
1 | 1 1
'
1 | 1 1
] ) ( [ ) (

+ + + + + +
+ =
k
Aug
k k k
Aug
k
Aug
k k k
Aug
k
R H P H H P K
(6)
k
' Aug
k k | k
Aug
k k | k
Q ) A ( P A P + =
+1

(7)
k | k
Aug
k
Aug
k k | k
P ) H K I ( P
1 1 1 1 1 + + + + +
=
(8)
Where
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 120
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
(

=
k
k Aug
k
U
X
X ,
(

=
u
k
x
k Aug
k
K
K
K ,
(

=
u
k
' xu
k
xu
k
x
k
k
P ) P (
P P
P
(

=
k n m
k k Aug
k
C
B A
A
0
,
'
0
(

=
m p
k
Aug
k
H
H
(

=
u
k
' xu
k
xu
k
x
k
k
Q ) Q (
Q Q
Q
Where the superscript Aug denotes the
augmented system state, I denotes the identity
matrix of any dimension and
n m
0 is a n m zero
matrix. It is clear from (4)-(8) that the
computational cost of the AUSKE increases with
the augmented state dimension. The reason for
this computational complexity is the extra
computation of
xu
k
P terms in each sample time k .
Therefore, if this term can be eliminated, one can
reduce the complexity of computational effort. In
this paper, we propose a new optimal two-stage
Kalman estimator without calculating the term of
xu
k
P explicitly. Therefore, the proposed scheme is
developed to reduce the computational cost of an
AUSKE by partitioning the equations (4)-(8) into
two subsystems [17].
1. DERIVATION OF THE OPTIMAL PARTITIONED
STATE KALMAN ESTIMATOR
The design of a new two-stage Kalman estimator
is described as follows. First, define a modified
input-free model and design a modified input-free
filter by ignoring the input term. Second, derive
an input filter to compensate the modified input-
free filter in order to minimize mean square error.
These two filters are used to build a new scheme,
which is equivalent to the AUSKE. The major
derivation is the relation between the
measurement residues of the two different filters.
One is the measurement residue of the input-free
filter, which does not consider unknown input
vector, and the other is the measurement residue
of the input filter. Based on the measurement
residues of the two filters, an input estimation
algorithm is derived using the minimum mean
square estimation technique [17].
The input-free model can be obtained by
ignoring the input term ( 0 =
k
U ) in (1) as below:
x
k k k k
W X A X + =
+1
(9)
Where the state vector of the input-free model is
denoted by
k
X . The input-free filter is just a
Kalman filter based on the model (9) and (3) as
below:
) X

H Z ( K X

k | k k k k k | k k | k 1 1 1 1 1 1 1 + + + + + + +
+ = (10)
k | k k k | k
X

A X

=
+1
(11)
1
1 1 1 1 1 1

+ + + + + +
+ = ] R ) H ( P H [ H P K
k
'
k
x
k | k k
'
k
x
k | k k
(12)
x
k
'
k
x
k | k k
x
k | k
Q ) A ( P A P + =
+1
(13)
x
k | k k k
x
k | k
P ) H K I ( P
1 1 1 1 1 + + + + +
= (14)
In the following, we propose an expression which
relates the state vector of the input model
k
X to
the state vector of the input-free model
k
X . The
state vector of the input model in (1) using the
state of the input-free model in (9) can be
calculated for each sample time [17]:
u
x u
x
W C B U C B X
W ] W C U C [ B X A
W U B X A X
0
1
0 0 1
1
0 0 1
0 0
1
0 1
1
0 0 0 0
0 0 0 0 0 1


+ =
+ + =
+ + =
(15)
Since
k
C is a nonsingular matrix,
0
U in (15)
replaced by (3). It is assumed in (15)
that
0 0
X X = , so
x x
W X A W X A X
0 0 0 0 0 0 1
+ = + = .
Following the same procedure, we can define and
derive the expression for
2
X by using (15):
u
u
u u
u
x u
x
W C B A
W C B C B A U C B C B A X
W C B A W C U C B C B A X
W C B A U B C B A X
W U B W C B U C B X A
W U B X A X
0
1
0 0 1
1
1
1 1
1
0 0 1 2
1
1 1
1
0 0 1 2
0
1
0 0 1 1
1
1 2
1
1 1
1
0 0 1 2
0
1
0 0 1 1 1
1
0 0 1 2
1 1 1 0
1
0 0 1
1
0 0 1 1
1 1 1 1 1 2
] [ ] [
] ][ [
] [
] [

+ + + =
+ + =
+ + =
+ + + =
+ + =

(16)
For an arbitrary sample time k , we will have

=
+ + + +
+ =
k
i
u
i i k k k k
W U M X X
0
1 1 1 1
e (17)
Where
1
0 0 1
1
1
3 2

+
=
= + =
C B M
,.... , k , C ] B M A [ M
k k k k k
(18)
1
0 0
1
0
0
1
1
0
1
2 1

=
= + =
C B ) A (
k ,.., , i , C ] B ) A ( [
j k
k
j
i i j k
i k
j
i i
H e
H e e

(19)
By using zero-mean property of
u
k
W and one-step
prediction logic from dynamic equation (17) the
predicted state is obtained as follow [17]:
1 1 1 1 + + + +
+ =
k k k | k k | k
U M X


(20)
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 121
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
The updated state for the input model can be
considered using the state of the input-free model
as below:
1 1 1 1 1 1 + + + + + +
+ =
k k k | k k | k
U N X

(21)
Where
1 + k
N in (21) must be calculated. Hence,
the relation between two innovation matrices
k
M
and
k
N needs to be determined. Suppose that the
updated state
1 1 + + k | k
X

obtains from the Kalman


filter framework:
) X

H Z ( K X

k | k k k k k | k k | k 1 1 1 1 1 1 1 + + + + + + +
+ =
(22)
By some manipulation and using (10) and (22), an
expression to relate
1 + k
N and
1 + k
M can be
obtained as below:
1 1 1 1
1 1 1 1
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1 1
+ + + +
+ + + +
+ + + + +
+ + + + +
+ + + + + +
=
=

+ =
=
k k k k
k | k k | k k k
k | k k k k k | k
k | k k k k k | k
k | k k | k k k
U M ] H K I [
] X

][ H K I [
] X

H Z [ K X

] X

H Z [ K X

U N

(23)
The equation, which relates
1 + k
N and
1 + k
M can
be obtained by comparing both sides of the above
equation;
1 1 1 1 + + + +
=
k k k k
M ] H K I [ N
(24)
Once the initial value of
k
M is chosen,
1 + k
M can
be recursively calculated by (18), and the
1 + k
N
can be obtained by (24) in each iteration. Note
that the
1 + k
K for the input filter is equal to
1 + k
K
for the input-free filter since the input term
assumes to be nonrandom. In addition, both filters
have the same covariance matrices. In other
words, the equality of filter gain for
k
X

and
k
X
in these filters arises from the same error
covariance matrices of
k
X and
k
X denoted
by
x
k | k
P . The innovations
1 + k
Z
~
and
1 + k
Z
~
as the
measurement residues of the input-free model and
the input model are defined, respectively [17];
k | k k k k
X

H Z Z
~
1 1 1 1 + + + +
= (25)
k | k k k k
X

H Z Z
~
1 1 1 1 + + + +
= (26)
The difference of these two innovations is defined
as:
] X

[ H Z
~
Z
~
Z
~
k | k k | k k k k k 1 1 1 1 1 1 + + + + + +
= = A
(27)
Using (27) and (20), we represent
1 + k
U in the
equation, which relates the input state to the
measurement residue as below:
1 1 1 1
1 | 1 | 1 1 1
~
~
]

[
~
+ + + +
+ + + + +
+ =
+ =
k k k k
k k k k k k k
Z U M H
Z X X H Z

(28)
It is clear from the input-free filter (10)-(14)
which is a standard Kalman filter, that the
measurement residue
k | k
Z
~
1 +
exists and easily
obtainable while the measurement residue
k | k
Z
~
1 +
is
actually not available. Note that (28) is in the
standard form
1 1 1 1 + + + +
+ =
k k k k
v _ H Z , and can be
viewed as an observation model of
1 + k
U . In (28),
the measurement residue

k | k
Z
~
1 +
is a non-zero mean
white random process since the input term
introduces a bias in the innovation
k | k
Z
~
1 +
. The
amount of this bias will supply the information
about the existence of input value. In contrast, the
measurement residue
k | k
Z
~
1 +
is a zero mean white
random process ( 0
1
=
+
] Z
~
[ E
k
) with covariance
matrix
z
k | k
P . In other words, we would like to
estimate
1 + k
U in order to minimize the error
covariance matrix
u
k
P
1 +

under the
constraint 0
1 1 1 1
= =
+ + + +
] X

H Z [ E ] Z
~
[ E
k | k k k k
,
or
k | k k
X

] X [ E
1 1 + +
= , yield unbiased estimation.
The desired form of the filtering solution for
estimating the unknown vector
1 + k
U is a
difference equation expressing
1 1 + + k | k
U

in terms of
k | k
U

and
1 + k
Z
~
. In the following, we derive a
recursive algorithm to estimate
1 + k
U in order to
minimize the error covariance matrix of the input
vector or
u
k
P
1 +
. Suppose that
k | k
U
~
1 +
denotes the
input state residue;
k | k k k | k
U

U U
~
1 1 1 + + +
=
(29)
Using (28) and zero-mean property of
1 + K
Z
~
, we
would like to obtain a recursive algorithm in the
form of Kalman filter as below [17]:
] U

M H Z
~
[ K U

] Z
~
Z
~
[ K U

k | k k k k
u
k k | k
k | k k
u
k k | k k | k
1 1 1 1 1 1
1 1 1 1 1 1
+ + + + + +
+ + + + + +
+ =
+ =
(30)
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 122
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
Using the residue of
1 + k
U defined in (29) and
using (30) and (28), gives
1 1 | 1 1 1 1 | 1
| 1 1 1 1
1 1 1 1 | 1
| 1 1 1 1 1
| 1 1
1 | 1 1 1 | 1
~ ~ ~
]

~
[
~
]

~
[

~
+ + + + + + +
+ + + +
+ + + + +
+ + + + +
+ +
+ + + + +
=
+
=

=
=
k
u
k k k k k
u
k k k
k k k k k
k k k
u
k k k
k k k k k
u
k
k k k
k k k k k
Z K U M H K U
U M H Z
U M H K U
U M H Z K
U U
U U U

(31)
The covariance matrix of
1 1 + + k | k
U
~
defined in (31)
gives
) (
) (
) (
) ( ) (
) (
1
'
1
'
1 | 1 1
| 1 1 1 | 1 1 1 1
| 1 1 1 1 1 | 1
1
'
1
'
1 | 1 1 | 1 1
'
1
'
1
'
1 | 1 1 1 1 | 1
1 | 1
' +
' +
'
' ' +
+ =
+ + + + +
+ + + + + + +
+ + + + + +
+ + + + + + +
+ + + + + + + +
+ +
u
k k k
zu
k k
u
k
zu
k k
u
k
u
k
uz
k k k k
u
k
u
k k k k
u
k
u
k
uz
k k
u
k k k
u
k k
u
k
z
k k
u
k
u
k k k
u
k k k k
u
k
u
k k
u
k k
K H M P K
P K K P M H K
P M H K K P
K H M P K P K
K H M P M H K P
P

(32)
Where the covariance matrix of
1 + k
Z
~
, denoted by
z
k | k
P
1 +
must be calculated. From (26):
1 1 1 1 1 + + + + +
+ =
k
'
k
x
k | k k
z
k | k
R H P H P
(33)
It is seen that the input state residue
k | k
U
~
1 +
is time-
correlated with the measurement residue of the
input model
1 + k
Z
~
, denoted by
zu
k | k
P
1 +
.
} U
~
Z
~
{ E P
'
k | k k
zu
k | k 1 1 1 + + +
= (34)
The extra computation of this cross covariance
matrix
zu
k | k
P
1 +
(which relates to
xu
k | k
P
1 +
) is the reason
for the computational complexity in the
augmented state methods. Therefore, if this term
can be eliminated, one can reduce the complexity
of computational effort. In the following, we
propose an expression which relates
zu
k | k
P
1 +

to
u
k | k
P
1 +
. Since the magnitude of the input term
1 + k
U in (20) is unknown, we can only use the
estimation of
1 + k
U

to modify the state vector of


the input-free model
k | k
X
1 +
to obtain the state
vector of the input model
k | k
X
1 +
. Therefore, we
rewrite (20):
k | k k k | k k | k
U

M X

1 1 1 1 + + + +
+ =
(35)
Using the equations (17), (3), (26) and (35), gives:
1
0
1 | 1 1 1 | 1 1
1
0
1
| 1 1 1 1 | 1 1 1
1 | 1 1 | 1 1
0
1 1 1 1
1 | 1 1
1 | 1 1 1 | 1 1 1
1
~
~
]

[ ]

[
]

[
] [
~
]

~
+
=
+ + + + + +
+
=
+
+ + + + + + +
+ + + + +
=
+ + + +
+ + +
+ + + + + + +
+
+
+ =
+
+ =
+ +
+ =
+ =
+ =
=

k
k
i
u
i i k k k k k k k k
k
k
i
u
i i k
k k k k k k k k k
k k k k k k k
k
i
u
i i k k k k
k k k k
k k k k k k k k k
k
V
W H U M H X H
V W H
U U M H X X H
V U M X H
W U M X H
V X H
V X X H X H Z
Z
e
e
e

(36)
1
0
1
| 1 1 1 | 1 1 | 1
~
~
~
+
=
+
+ + + + + +
+
+ =
k
k
i
u
i i k
k k k k k k k k k
V W H
U M H X H Z
e
(37)
It should be noted that the term
1 1 1 + + +
+
k k | k k
V X
~
H
in (37) is not equal to
1 + k
Z
~
in (16). Using (37),
the cross covariance matrix
zu
k | k
P
1 +
can be
calculated:
u
k k k k
k k k k k k
k k k
k
i
u
i i k
k k k k k k k
k k k
zu
k k
P M H
U U E M H
U V W H
U M H X H E
U Z E P
| 1 1 1
'
| 1 | 1 1 1
'
| 1 1
0
1
| 1 1 1 | 1 1
'
| 1 1 | 1
}
~ ~
{
]
~
)
~
~
[(
]
~ ~
[
+ + +
+ + + +
+ +
=
+
+ + + + +
+ + +
=
=
+
+ =
=

e

(38)
Where ] U
~
X
~
[ E
'
k | k k | k 1 1 + +
, 0
1
0
=
+
=

] U
~
W [ E
'
k | k
k
i
u
i i
e
and 0
1 1
=
+ +
] U
~
V [ E
'
k | k k
are equal to zero. One
important property of the optimal estimation of
1 + k
U

is that the input residue


k | k
U
~
1 +
must be
orthogonal to
1 + k
Z ,
1 + k
V and
u
k
W or any linear
function of
1 + k
Z [17].
u
k | k k k
zu
k | k
P M H P
1 1 1 1 + + + +
=
(39)
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 123
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
Substitute
zu
k | k
P
1 +
in (32) the covariance matrix of
1 1 + + k | k
U

becomes [17]:
) (
) (
) (
) ( ) (
) (
1
'
1
'
1 | 1 1 1 1
| 1 1 1 1
1
'
1
'
1 | 1 1 1 1
| 1 1 1 1 1
'
1
'
1 | 1
1
'
1
'
1 | 1 1 | 1 1
1
'
1
'
1 | 1 1 1 1 | 1
1 | 1
'
+
' +
'
' ' +
' + =
+ + + + + + +
+ + + +
+ + + + + + +
+ + + + + + + +
+ + + + + + +
+ + + + + + + +
+ +
u
k k k
u
k k k k
u
k
u
k k k k
u
k
u
k k k
u
k k k k
u
k
u
k k k k
u
k
u
k k k
u
k k
u
k k k
u
k k
u
k
z
k k
u
k
u
k k k
u
k k k k
u
k
u
k k
u
k k
K H M P M H K
P M H K
K H M P M H K
P M H K K H M P
K H M P K P K
K H M P M H K P
P

(40)
u
k k k k
u
k
u
k k k
u
k k
u
k
z
k k
u
k
u
k k k
u
k k k k
u
k
u
k k
u
k k
P M H K
K H M P K P K
K H M P M H K P
P
| 1 1 1 1
1
'
1
'
1 | 1 1 | 1 1
1
'
1
'
1 | 1 1 1 1 | 1
1 | 1
2
) ( 2 ) (
) ( 3
+ + + +
+ + + + + + +
+ + + + + + + +
+ +

' ' +
' +
=

(41)
The second and the third terms of the above equation
can be considered as ) K ( W W K
u
k
u
k
' '
+ + 1 1
, where
z
k | k
'
k
'
k
u
k | k k k
P H M P M H W W
1 1 1 1 1 1
3
+ + + + + +
+ = ' (42)
Since the covariance matrices
u
k | k
P
1 +
and
z
k | k
P
1 +
are
symmetric, we can find a decomposition in the
form of W W ' , for appropriate matrix W , then
D D D W K D W K P
P
u
k
u
k
u
k k
u
k k
' ' +
=
+ + +
+ +
] ][ [
1 1 | 1
1 | 1

(43)
Comparing (41) and (43), the term D is defined
as below:
1
1 1 1
2

+ + +
' = ) W ( H M P D
'
k
'
k
u
k | k

(44)
The minimum
u
k | k
P
1 1 + +
is obtained by setting
W K
u
k 1 +
equal to D. Therefore,
1
| 1
'
1
'
1 | 1 1 1
'
1
'
1 | 1
1 '
1
'
1 1 | 1 1
] 3 [
2
) ( 2

+ + + + + +
+ + +

+ + + + +
+
=
' =
z
k k k k
u
k k k k
k k
u
k k
k k
u
k k
u
k
P H M P M H
H M P
W W H M P K

(45)
The minimum error covariance
u
k | k
P
1 1 + +
is
obtained:
u
k k k k
u
k
u
k k k k
z
k k k k
u
k k k k k k
u
k k
u
k k
k k
u
k k k k
u
k k
u
k k
u
k k
u
k k
P M H K I
P M H P H M
P M H H M P P
H M P W W H M P
P D D P
P
| 1 1 1 1
| 1 1 1
1
| 1
'
1
'
1
| 1 1 1
'
1
'
1 | 1 | 1
'
1
'
1 | 1
1 '
1
'
1 | 1
| 1 | 1
1 | 1
] 2 [
]
3 [ 4
] 2 [ ) ]( 2 [
+ + + +
+ + +

+ + +
+ + + + + + +
+ + +

+ + +
+ +
+ +
=
+
=
' '
= ' =

(46)
Based on (2), we have
u
k
'
k
u
k | k k
u
k | k
Q C P C P + =
+1
(47)
If the value of
u
k
K
1 +
given by (45) substitute in
(30), the estimation of
1 + k
U

leads to a minimum
error covariance [17].

IV. CONCLUSION
In this article, we introduce Cloud Computing and
perusal about influences of it on the processes of these
days. Cloud computing is combination of various
computing entities, globally separated, but
electronically connected. As the geography of
computation is moving towards corporate server
rooms, it bring more issues including security, such as
virtualization security, distributed computing,
application security, identity management, access
control and authentication. However, strong user
authentication is the paramount requirement for cloud
computing that restrict illegal access of cloud server.
In cloud computing, cloud providers can offer cloud
consumers two provisioning plans for computing
resources, namely reservation and on-demand plans. In
general, cost of utilizing computing resources
provisioned by reservation plan is cheaper than that
provisioned by on-demand plan, since cloud consumer
has to pay to provider in advance. With the reservation
plan, the consumer can reduce the total resource
provisioning cost.
Although creating a Cloud Computing architecture that
is scalable and is usable for sharing all kind of
resources, has so many problem and complexions, but
it can be usable for optimization and removing all IT
requirements. These days lots of technologies migrate
from traditional systems into cloud, and cloud
computing has developed and used in so many
countries. These countries are using cloud computing
in many of industries with different applications and
also the range of using cloud computing is increasing
in different countries and with different applications.
Although, there is some worry about security in cloud
computing, but the number of persons that save their
personal information in servers of third company for
example Google, is increasing. We presented our new
JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 124
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617
idea for improving its security; using of two stage
Kalman estimator for predicting and updating data
about the amount of crowd on different cloud sections
and also improving the security level of such network.
With regard to lots of cloud computing advantages,
specially, costs reduction of implementation in large
scale, investing capital is increasing in this filed. Cloud
Computing is advancing with fast rate and also it will
be complete with little deficiencies rather than other
technologies. It is predict that Cloud computing is the
basic platform for IT in next 20 year [16].

REFERENCES
[1] David C. Wyld; the cloudy future of government
IT: cloud computing and the public sector around
the world, IJWesT, Vol. 1, Num. 1, Jan. 2010.
[2] Jean-Daniel Cryans, Alain April, Alain Abran;
criteria to compare cloud computing with current
database technology, R. Dumke et al. (Eds.):
IWSM / MetriKon / Mensura 2008, LNCS 5338,
pp. 114-126, 2008.
[3] Anil Madhavapeddy, Richard Mortier, Jon
Crowcroft, Steven Hand; multiscale not
multicore: efficient heterogeneous cloud
computing, published by the British Informatics
Society Ltd. Proceedings of ACM-BCS Visions of
Computer Science 2010.
[4] Harold C. Lim, Shivnath Babu, Jeffrey S. Chase,
Sujay S. Parekh; automated control in cloud
computing: challenges and opportunities,
ACDC09, June 19, Barcelona, Spain.
[5] N. Sainath, S. Muralikrishna, P.V.S. Srinivas; a
framework of cloud computing in the real world;
Advances in Computational Sciences and
Technology, ISSN 0973-6107, Vol. 3, Num. 2,
(2010), pp. 175-190.
[6] Kyle Chard, Simon Caton, Omer Rana, Kris
Bubendorfer; social cloud: cloud computing in
social networks
[7] G. Bruce Berriman, Eva Deelman, Paul Groth,
Gideon Juve; the application of cloud computing
to the creation of image mosaics and management
of their provenance,
[8] Roy Campbell, Indranil Gupta, Michael Heath,
Steven Y. Ko, Michael Kozuch, Marcel Kunze,
Thomas Kwan, Kevin Lai, Hing Yan Lee, Martha
Lyons, Dejan Milojicic, David OHallaron, Yeng
Chai Soh; open cirrus TM cloud computing
testbed: federated data centers for open source
systems and services research
[9] Rajkumar Buyya, Chee Shin Yeo, Srikumar
Venugopal, James Broberg, Ivona Brandic; cloud
computing and Emerging IT platforms: Vision,
Hype, and Reality for delivering computing as the
5th utility
[10] Lamia Youseff, Maria Butrico, Dilma Da Silva;
toward a unified ontology of cloud computing
[11] Daniel A. Menasce, Paul Ngo; understanding
cloud computing: experimentation and capacity
planning; Proc. 2009, Computer Measurement
Group Conf. Dallas, TX. Dec. 2009.
[12] Won Kim; cloud computing: today and
tomorrow; JOT, Vol. 8, No. 1, Jan-Feb 2009.
[13] Richard Chow, philippe Golle, Markus Jakobsson,
Elaine Shi, Jessica Staddon, Ryusuke Masuoka,
Jesus Molina; controlling data in the cloud:
outsourcing computation without outsourcing
control; CCSW09, Nov. 13, 2009, Chicago,
Illinois, USA.
[14] Bo Peng, Bin Cui, Xiaoming Li; implementation
issues of a cloud computing platform; Bulletin of
the IEEE computer society technical committee on
data engineering.
[15] Daniel Nurmi, Rich Wolski, Chris Grzegorczyk,
Graziano Obertelli, Sunil Soman, Lamia Youseff,
Dmitrii Zagorodnov; the eucalyptus open-source
cloud computing system.
[16] Mehdi Darbandi, Hoda Purhosein; Perusal about
influences of Cloud Computing on the processes
of these days and presenting new ideas about its
security, Int. IEEE Conf. Azerbaijan, Bakku.
[17] H. Khaloozade, M. Darbandi; optimal partitioned
state Kalman estimator Int. IEEE Conf. Harbin,
China.










Mohsen Panahi











Mehdi
Darbandi







Mohammad
Abedi










Ali Hamzenejad

















JOURNAL OF COMPUTING, VOLUME 4, ISSUE 9, SEPTEMBER 2012, ISSN (Online) 2151-9617
https://sites.google.com/site/journalofcomputing
WWW.JOURNALOFCOMPUTING.ORG 125
2012 Journal of Computing Press, NY, USA, ISSN 2151-9617