Content uploaded by Hiral Patel
Author content
All content in this area was uploaded by Hiral Patel on Sep 30, 2023
Content may be subject to copyright.
2017 Nirma University International Conference on Engineering (NUiCONE)
978-1-5386-1747-2/17/$31.00 ©2017 IEEE
A Novel Approach for Securely Processing
Information on Dew Sites (Dew Computing) in
Collaboration with Cloud Computing
An Approach toward latest research trends on Dew Computing
Hiral Patel
Department of Computer Engineering
Sankalchand Patel college of Engineering
Visnagar, Gujarat, India
hmpatel.ce@spcevng.ac.in
Krunal Suthar
Department of Computer Engineering
Sankalchand Patel college of Engineering
Visnagar, Gujarat, India
krunal_bece@yahoo.co.in
Abstract—A computer user now-a-days requires a freedom
from managing data locally and wants that the data must become
available anytime and anywhere. Here the term Cloud comes into
existence which provides various services to its user without
taking a burden as well as with cheaper cost. This gives lots of
benefits to user but oppositely it is also suffer from some of
burning issues like confidentiality, availability, continuous
network connection requirement, integrity and many more. The
user data available on Cloud system and to process that data
securely a provider need to process this data without displaying
identity of the user. Secondly the user must be able to work with
open environment even the network connection is not available.
First issue is solved using the anonymization techniques and for
second issue we used an important new term that’s Dew
computing. The Dew computing together with Cloud makes a
Cloud-Dew architecture which united works to solve above
mentioned issues. We are among few researchers who proposed a
scheme which solved important issue related to privacy of user
information in Cloud-Dew Computing architecture that may be a
one of the important open platform computing trends in the
future.
Keywords—Cloud Computing, Dew Computing, Security,
Anonymization, Cloud-dew Architecture.
I.
I
NTRODUCTION
The way to store and retrieve personal as well as
commercial information has been totally refined due to
enormous innovations in today’s world. With fast growth of
Internet, users can access their data anywhere in the world
without carrying the data on a physical device. In early 2000,
the new notion was introduced titled as “cloud computing”,
which is the technology of storing and accessing data as well
as applications over an Internet network. Cloud computing
uses network of shared large pools of systems, resources and
servers. With the concept of pay on use, cloud computing
architecture allows the client to procure services at different
level of abstraction such as Platform as a Service, Software as
a Service and Information as a Service, depending upon their
requirements [1]. It gives users the freedom from location
bounding as users can access services everywhere with the
help of availability of Internet and a standard web browser,
allowing working on a single project from multiple Geo-
distributed workplaces.
Considerable extent of data processing nearby ourselves is
accomplished on the lowermost plausible computing level,
directly linked to the physical surroundings. Data processing
devices we can find in the industry as well as industrial
products. These devices which are at the physical edge of
computing are the root of the Dew Computing.
As we know, Cloud computing requires Internet
connection. The evident detriment of keeping data using the
Cloud utility is forfeiting access in the absence of Internet
connection. Because all resources are far from user’s premises
and out from user’s control, if an Internet connection is lost,
the user will not be able to access the user’s own data. To
eliminate this problem one more new concept comes into the
picture known as “Dew computing”
For the Preprocessing, conversion of user data into proper
format is required before sending them on the Cloud. There
may be issue related to leaking the user identity when some
processing done on the cloud premises or even after fetching
the data on local Dew site which is a website hosted on user's
local computer. Authors at [17] have mentioned that data can
be secured on cloud using obfuscation techniques so that it
cannot be misused even by cloud providers. Even though
database technology is matured, some challenges related to
privacy of user information exist when databases are used in
dew computing. To solve these issues we need to use
anonymization techniques which help to hide the user identity
when commonly processed large amount of data of different
user on common platform like Cloud-dew environment.
II. B
ACKGROUND THEORY
A. Dew Computing
Dew Computing is a prototype whose objective is to
wholly grasp the abilities of personal computers as well as
cloud services. In such kind of archetype, organization of soft-
wares on a personal computer is based on the Cloud-dew
Architecture which offers lavish utilities independent and
collaborates with cloud services. Dew Computing is the future
direction of on-premises computer applications [6]. “Dew
computing is a method where the on-premises computer
provides utilities which are independent and also collaborative
with cloud services. The goal of Dew computing is to fully
realize the potentials of on-premise computers and cloud
services” [2].
The independence feature encourages using on-premises
resources as far as possible before sending requests to cloud
services to fully understand the power of on-premises
computers[2].Meaning of Collaboration is exchange of
information with cloud services automatically during dew
computing application's operation. Collaboration may
involves synchronization, correlation, or any other type of
inter operations. The collaboration feature realizes the
potentials of cloud services by promoting the use of cloud
services together with on-premises computers. Independence
suggests inherently distributed nature of application whereas
collaboration suggests inherently connected nature of dew
computing application [2].
B. Cloud-Dew Architecture
The task of making synchronization between data on cloud
and local computer is very trivial in case of complex data. In
[4] author's architecture follows the conventions of Cloud
architecture, in addition to Cloud servers, there are dew
servers which are situated on the native system and act as a
buffer between the local user and the Cloud servers, also
abstain the enigma of data becoming out of synchronization.
This dew server would essentially host scaled-down variants
of websites, full of per-downloaded contents, which the user
could access without Internet connection [6]. Now the website
hosted on user's local computer is known as “Dew site”. The
major two functions performed by dew server and its related
databases are: providing the client services same as services
provided by the Cloud server and another is, maintaining
synchronization between databases of dew server and cloud
server.
Cloud-dew architecture is an extension of the client- server
architecture [3]. A dew server has the following features [5]:
(1) A dew server is a light weight web server, which is
able to serve only one user
(2) A dew server can store only user’s data because a dew
server is very small like a drop of dew while a cloud server is
very big like a real cloud.
(3) A dew server is week as real drop of dew because a
dew server's data vanishes easily due to hardware failure,
infection of virus etc.
(4) Because cloud can provide all the necessities, after
disappearance dew will come out again, similarly vanished
dew server can be entertained once again because of copy of
all dew server data in the cloud servers.
(5) A dew server is running on the local computer, so it is
available with or without an Internet connection.
This architecture can furthermore be employed to make
available websites off line. Suchlike system can diminish the
overhead of Internet data for an organization having weak
Internet connectivity. Many functions like displaying files or
images, playing audio or video would be possible without
Internet connection but provided that data had been
synchronized to the "Dew site" from the web over the last
connection interval.
C. Anonymization [13]
Since Dew servers will likely come across sensitive user
informations such as browsing activity, medical histories,
messages or any personal sensitive information. So, preserving
user privacy is major concern when processing information
mentioned above on dew site because of there is no direct
control of cloud server on dew server. One of the methods to
achieve privacy is to remove all personally identifiable
information based on key attributes available. This is achieved
using anonymization techniques. Anonymization is a type
of classified information whose intent is to achieve privacy
protection [15].There is three major types of Anonymization
techniques.
K Anonymity [11] [12]
This model is developed because indirect identification of
record from public database. This model protect against
“identity discloser”. It used transformation techniques like :
generalization: replace the original value by a semantically
consistent but less specific value and suppression: one can
hide some sensitive attribute of the record.
K-Anonymity gets failure in providing privacy if there is
absence of diversity of sensitive values in an equivalence class
as well as when attacker has any background knowledge.
Below Fig.1 shows the example of k-
Anonymization.
1
Fig. 1. K- Anonymization Example [19]
L Diversity [14][16]
L-diversity is a form of group based anonymization that is
used to preserve privacy in data sets. The L-diversity model is
an extension of the k-anonymity model which reduces the
granularity of data representation using techniques including
generalization and suppression. The L-diversity model handles
some of the weaknesses in the K-anonymity model. K-
anonymization is insufficient to prevent attribute discloser.
But L-diversity has also some bounds.
Sometimes L-diversity may become arduous and unessential
to accomplish when having single sensitive attribute or having
an equivalence class consisting of only negative records.
l-diversity is deficient to preclude attribute disclosure and also
it does not take into consideration the semantic meanings of
sensitive values. Below Fig. 2 shows L-Diversity.
2
Fig. 2. L-Diversity Example [19]
T Closeness [15]
K-anonymity prevents identity disclosure but it is not
capable for prevention of attribute disclosure. So, to figure out
that problem l-diversity is used which requires that each
equivalence class must have at least l values for each sensitive
attribute. But as we discussed above l-diversity also has
certain bounds. Hence, T-closeness comes into picture-
closeness needs that the distribution of a sensitive attribute in
any equivalence class is near to the distribution of a sensitive
attribute in the entire table. Below Fig. 3 depict T-closeness
example.
3
Fig. 3. T Closeness Example [19]
Below Fig. 4 indicates summary of available approaches in
anonymization techniques and we must say t-closeness works
better than other two approaches.
4
Fig. 4. Anonymization Techniques Summary [18]
III. L
ITERATURE
R
EVIEW
Authors at [7] proposed a security model based on K-
Anonymity and e-differential. Authors have focused on high
computational efficiency and about loss of information which
is satisfy using k-anonymity. Authors also argued that the
proposed algorithm for k-anonymity is generally based on
generalization and suppression. They have focused on
problem related to combining generalization and
suppression with its drawback. As the data processed partially
it is very difficult to suppress the data because of inability to
deal with continuous micro data. Their main goal behind the
proposal is to find natural and unified way to achieve k-
anonymity through microaggragation for different types of
attributes.
Authors at [9] find the limitation of Implementation of
cloud storage due to privacy troubles when inserting and
fetching susceptible information. Authors gave solution for
privacy of user information by dividing tables which contains
identity and sensitive information. They have proposed
scheme which consider different related variables like
computation cost for local processing, communication cost of
local implementation, response time, independent variable etc.
The research mainly used to solve user defined problem
regarding query processing in cloud based database
applications. Authors have also argued about the efficient
technique related to cloud based database applications called
privacy safety anonymity which is better than the other
proposed methodology. Authors also proposed solutions
which work on outsourced database by writing queries without
affecting on the privacy of each individual. Authors have
concluded that this is first approach that they proposed in
which no need to store up original database and may not use
encryption. They also discuss result and performance analysis
which show that the proposed technique is efficient and tough
under various parameter settings.
In the proposed model authors at [10] use two protocols
for solving issue of confidential databases using suppression-
based and generalization-based k-anonymous. These two
protocols are based on assumption on the well known
cryptographic primitives. Authors only provide the theoretical
analyses using assumptions. They have tried to provide proof
related to soundness and efficiency through experimental
results. Author also argued that when new tuple is inserted,
proposed scheme provides enough security on k-anonymous
database which ensure retention of anonymity. They conclude
that the query sent by the user or patient on E- health cloud,
security at data provider’s side cannot be dishonoured.
At [4] authors gives idea about existing cloud-dew
architecture, two kinds of URLs are considered: regular URL
such as https://www.test.com and local URL such as
https://mmm.test.com. On user's local computer the website is
hosted, which is known as “dew site”. Here mmm can be used
to indicate dew site whereas www is used to indicate website.
All names of the dew sites that user wants on his dew server
must be placed in the host file ,which is available in almost all
operating system and maps host names to IP addresses .A dew
server can be accessed with the help of local host. When a user
enters URL in the navigation bar of the browser, if a URL is
correspond to a website(i.e. beginning with www)then
browser follows the steps to map the domain name to the IP
address using DNS server and display website content and if a
URL is correspond to a dew site(i.e, beginning with
mmm)then dew server check the existence of domain name in
host file. If host file does not contain requested domain name
then dew server will send request for script and database of
requested website to the remote domain(cloud server).when
the requested is approved by remote domain then script and
database of website will be integrated in dew server. Then
local URL will be mapped to the local-host. Dew server will
then find target host name by using environmental variable.
The URL request will be then redirected to the corresponding
dew site script. User will then perform operation on dew
site[5]. synchronization with cloud server will started on the
availability of the Internet connection. Now, to perform
synchronization between content of dew site and website, user
have to logged into website .Once the user logged into
website, after doing internally mapping of user id of dew site
and user id of website , a link is created between user on dew
site with website and synchronization will be started
automatically.
IV. P
ROPOSED
M
ETHODOLOGY
The proposed methodology works in four different levels
as shown in below Fig 5.
5
Fig 5. Proposed model (Basic Diagram)
A. User Registration
In User registration module, user have to fill up various
fields such as username, password, mobile number, email ID
etc. If any of the field is not filled correctly then an error
message will appear. User must have to register with the cloud
before performing any operation.
B. Login
User have to enter correct user name and password for
login. If any problem is with user name or password then error
message will be show to enter correct details once again. After
successful login user is able to create database. Once the user
uploaded successfully then message will be display that you
have successfully uploaded the file.
C. Data Upload
In this phase, various data providers upload information on
the cloud may be related to Salary, Medical information
,personal information etc. by specifying sensitive attributes or
quasi-identifiers. This helps in preventing leaking of user
identity while some processing is done on user information .
D. Data Download
We can take advantage of Dew computing for
downloading data which will store data once on dew server
and user can execute the query locally many times even
though network is not available after downloading.
E. Data processing
Now data will process and the result will be checked by
anonymization technique. If anonymity retains then result will
be display by hiding sensitive attributes otherwise proper
message about not a success in achieving anonymity in query
execution will be display. Now, on availability of Internet
connection modified database will be synchronized
automatically with the database on cloud server.
Below three charts in Fig. 6 , Fig. 7 and Fig. 8 give the
detailed description about how data provider sends detail on
cloud, how user can download required data on dew server
and how execution of query performed on dew server and
synchronization of modified content on cloud server
respectively.
6
Fig 6. Proposed model (Detailed Chart1)
7
Fig 7. Proposed model (Detailed Chart2)
8
Fig 8. Proposed model (Detailed Chart3)
V. I
MPLEMENTATION
&
R
ESULT
D
ISCUSSION
We have used core i3 processor with windows 8 operating
system and 2GB RAM for implementation. We have used Net
beans IDE for the development and used JAVA as a
programming Language.
9
Fig 9. Implementation (Attribute based identification)
The above Fig. 9 gives the basic screen shot of implemented
model which is based on the data available and by using
different group of Attributes (single, two etc.).The percentage
of hiding of information using anonymization is changed.
W have taken different types of dataset in Cloud-dew
environment and when we send a query then the percentage of
losing of data is shown in below analysis. Below Fig. 10
shows the comparison of three different methods of
anonymization.
The below Fig. 11 shows that the time requires in processing
the data on dew site as well as if the same number of query
processed on the Cloud server.
10
Fig 10. Analysis (Different Dataset)
11
Fig 11. Analysis (Dew and Cloud Computing)
CONCLUSION
The Cloud computing become very useful for all kind of
users now- a-days but availability of data on remote site as
well as the securely processing a data becomes difficult in
open environment which may reduce the adoption of Cloud
for lots of users. The continuous internet requirement is also
an important concern while adopting the Cloud concept.
So here we aim to propose an architecture with which user is
able to do task offline on Dew sites on local premises in
absence of Internet connection and when net connection is
available synchronization is done with cloud server. User is
also able to securely process the important information in
Cloud-Dew environment which ensure that the user identity
will be hidden from other users. We have provided the
detailed charts related to proposed scheme and also provided
implementation details. From the analysis we must say that the
identity of the user is not disclosed using a T-closeness
method while processed on dew site or on cloud platform. We
also show the comparison between different number of query
processing time taken by Dew and the Cloud sever with
anonymization techniques and result show that the Dew
computing work effectively better and saves lots of user
time.The Cloud-dew architecture related issue are new and
fully packed of challenges. For the future enhancement our
aim is to use efficient synchronization mechanism to deal with
issue of synchronized Dew sites data with the Cloud data.
R
EFERENCES
[1] Evolution of Cloud to Fog Computing,
https://blog.rankwatch.com/evolution-of-cloud-to-fogcomputing/
[2] Yingwei Wang,Definition and Categorization of Dew Computing. Open
Journal of Cloud Computing (OJCC), Vol.3,Issue 1(2016)3.
[3] Karolj Skala, Davor Davidovi, Enis Afgan, Ivan Sovi, Zorislav
Sojat,Scalable Distributed Computing Hierarchy:Cloud, Fog and Dew
Computing. Open Journal of Cloud Computing (OJCC),2,1(2015).
[4] Yingwei Wang,Cloud-dew architecture. Int. J. Cloud Computing,Vol.4,
Issue 3,pp. 199—210 (2015).
[5] Yingwei Wang,Cloud-dew architecture: realizing the potential of
distributed database systems in unreliable networks .In International
Conference on Parallel and Distributed Processing Techniques and
Applications (PDPTA), pp. 85--89 (2015).
[6] Andy Rindos,Yingwei Wang,Dew computing:the Complementary Piece
of Cloud Computing. In 2016 IEEE International Conferences on Big
Data and Cloud Computing(BDCloud),Social Computing and
Networking(SocialCom),Sustainable Computing and
Communications(SustainCom),pp. 15-20 (2016)
[7] David Edward Fisher,Shuhui Yang,Doing More with the Dew:A New
Approach to Cloud-Dew Architecture. Open Journal of Cloud
Computing (OJCC),vol.3,Issue 1(2016).
[8] J. Domingo-Ferrer and V. Torra. “Ordinal, continuous and heterogeneous
k-anonymity through microaggregation. Data Mining and Knowledge
Discovery,” 11(2):195-212, 20011.
[9] V. R.Patil and A.C.Lomte,“Challenges toward Achieving Privacy and
Secure Searchable Outsource cloud data Storage Services”,International
Journal of Advanced Research in Computer Science and Software
Engineering,Volume 3, Issue 11, pp. 400-403, 2013.
[10] Alberto Trombetta, Wei Jiang “Privacy-Preserving Updates to
Anonymous and Confidential Databases” in IEEE Transaction on
Dependable and Secure Computing, VOL. 8, NO. 4, July 2011.
[11] T. M. Truta and B. Vinay. “Privacy protection: p-sensitive k-anonymity
property”. In Proceedings of the 22nd International Conference on Data
Engineering Workshops, the Second Intenational Workshop on Privacy
Data Management (PDM’06), page 94, 2006.
[12] J. Domingo-Ferrer. “A critique of k-anonymity and some of its
enhancements.” In: Proc. of ARES/PSAI 2008, IEEE Computer Society,
[13] A. Sakhare and S. Ganar, “Anonymization: A Method to Protect
Sensitive Data in Cloud”,In International Journal of Scientific &
Engineering Research, Volume 25, Issue 2, pp. 1-11, 2013.
[14]
Chaitra.S , Narasimha Murthy M S “E-Health Care Solutions Using
Anonymization” International Journal of Advanced Research in
Computer and Communication Engineering Vol. 4, Issue 5, May 2015 to
IJARCCE DOI 10.17148/IJARCCE.2015.4591 417.
[15] Anu Rinny Sunny “Privacy Preserving of Data Using K-
Anonymisation And T-Closeness” International Journal for
Research in Applied Science & Engineering (IJRASET)
[16] Ashwin Machanavajjhala , Johannes Gehrke , Daniel Kifer,“ℓ-Diversity:
Privacy Beyond k-Anonymity” on Department of Computer Science,
Cornell University.
[17] Suthar K., Patel J “EncryScation: A Novel Framework for Cloud IaaS,
DaaS security using Encryption and Obfuscation Techniques” In 5
th
Nirma University International conference on Engineering(NUiCONE)
Dec 2015.
[18] Vijayalakshmi, V, A S Arunachalam, Mca, R Nandhakumar and M
Tech. “Mining Social Media-Utility Based Privacy Preservation.” In
International Journal of Computer Science and Information
Technologies, Vol. 5,Issue 4, 2014, 5480-5485 2014.
[19] https://www.cs.purdue.edu/homes/li83/papers/icde_closeness.pdf