International Journal of Innovative Technology and Exploring Engineering (IJITEE)
ISSN: 2278-3075, Volume-8 Issue-10, August 2019
2755
Published By:
Blue Eyes Intelligence Engineering
& Sciences Publication
Retrieval Number J95690881019/2019©BEIESP
DOI: 10.35940/ijitee.J9569.0881019
Abstract : Mobile Grid is a crossbreed technology formed by
amalgamation of the two prominent technologies namely mobile
technology and grid technology that enable sharing and collaboration of
mobile resources cooperatively, transparently, efficiently, reliably and
securely. Mobile Grid considers the mobility issues and overcomes the
constraints and deficiencies in both the technologies. However, this
heterogeneous, dynamic and open mobile grid network is more prone to
malicious and selfish nodes inside and outside the network. Hence, a
vigorous security mechanism is needed that considers different security
threats and pro-vide different levels of security services. Here, we propose
one such preventive security service based on Trust Management. The
proposed trust management service uses a novel fuzzy lattice approach
for trust estimation of the nodes in the network. A node with high trust
value is allowed to participate in the network. A malicious node having
low trust value is prevented from performing the task. A fuzzy lattice
approach can compute incrementally the same intervals in the training
data independent of the order of presentation within a short period.
Experimental analysis of the fuzzy lattice approach shows that the
proposed approach outperforms most of the existing approaches based on
fuzzy logic.
Keywords: Mobile Gird, Trust Management, Fuzzy Lattice
I. INTRODUCTION
In order to meet the fluctuating and on-demand resources
requirements one of the most promising technology has been
in the forefront in the form of Grid Computing. Grid
computing allows to share and allocate heterogeneous and
distributed resources dynamically. This results in an open and
dynamic environment providing computational and storage
resources in the form of grid services. The grid service
providers need to ensure a secure grid environment to the
users of the remote resources for executing their tasks
remotely and storing the data on the remote storage resources
securely.
With the exponential growth of wireless electronic devices
such as smart phones, PDA, laptops etc. along with the high
speed internet many recent advents have been done by the
researchers and industry to enrich the new comping paradigm
of mobile computing. Mobile computing allows collaboration
of mobile devices having limited resources such as battery,
processor, input/output interfaces and instability in data
transfer to solve a common problem. Providing security of
such limited and precious resources in mobile devices which
are being shared in a highly dynamic, open and heterogeneous
environment is a challenging problem.
The Mobile Grid [1] is a crossbreed technology incorporating
grid of mobile devices thus addressing mobility issues and
providing mobility to the resources and users in a continuous,
Revised Manuscript Received on August 05, 2019.
Grantej Vinod Otari, Department of Computer Science & Engineering
Shivaji University Kolhapur, India
Dr. Vijay Ram Ghorpade, Department of Computer Science &
Engineering Bharati Vidyapeeth's College of Engineering Kolhapur, India
Dr. Sachin Harakhchand Dhanani, Department of Mathematics
K.I.T.’s College of Engineering, Kolhapur, India
transparent, secure and effective manner. This allows us to
form a self-organized grid system consisting of an underlying
ad-hoc network of mobile devices interconnected by wireless
network and constructing random and dynamic network
topology. Thus the security infrastructure in the mobile grid
system should deal with various aspects of security issues
both in grid computing and mobile computing.
In the mobile grid network, every node plays the dual role as
client and server. Thus mobile grid resources are exposed to
distributed and open dynamic environment. However, such
mobile grid networks are extremely prone to malicious
participants dispersing false contents causing unrecoverable
security threat to the system. A viable solution is to develop a
trust model which provides a mechanism to establish a trusted
relationship between the participating resources and allowing
them to share the task and data les securely collaboratively. In
the trust model, every peer assesses every other peer in the
network after each trans-action. Then a peer selects the
trustworthy peer for further transaction based upon its past
transaction experiences.
Evaluating trustworthiness of a peer in the mobile grid is a
complex problem as trust is a linguistically fuzzy concept. To
solve this complex problem of trust calculation fuzzy logic is
a good alternative solution. Also it has been observed that all
the peers in the network are not always cooperative and may
send false feedback to disrupt the reputation of the peers and
contribute to the errors in global trust calculation. Thus a
robust trust estimation model is needed that detects malicious
peers and check the credibility of the recommendations
received from such peers. In addition the trust model should
also deal with the estimation of trust of the newly joined node
in the network.
In this paper, we have designed a novel trust management
system based on a fuzzy lattice approach. The proposed
model uses multiple attributes of the mobile node to evaluate
the direct trust value. These input attributes indicate the
capability of the node to perform the specific task based on
the currently available resources and its previous
performance. The trust model then estimates indirect trust by
collecting the recommendations from the neighbors in the
network and considering the credibility of the recommenders.
Finally, the obtained direct trust value and indirect trust value
is aggregated to compute the global trust of a node.
II. RELATED WORK
There exists a vast and diverse literature for development of
trust model. Numerous possible approaches and measures are
used for trust calculation. Some approaches use continuous
values to measure the trust, while some methods use discrete
values. Some models are based
on probabilistic approach
whereas some others use
threshold based approach.
A Fuzzy Lattice System to Trust Management in
Mobile Grid
Grantej Vinod Otari, Vijay Ram Ghorpade, Sachin Harakhchand Dhanani