.Privacy Preservation And Data Security In Location Based Services.pdf

Text-only Preview

Journal of Science and Technology
Volume 1, Issue 1, December 2016, PP 13-16
www.jst.org.in
www.jst.org,in 13 | Page
Privacy Preservation And Data Security In Location Based
Services
Jeevan B.S1, Dr. Naveen N.C2
1(PG Scholar/ Dayananda Sagar College of Engineering, Bangalore, India)
2(Professor and Head, Department of ISE / Dayananda Sagar College of Engineering, Bangalore, India)
______________________________________________________________________________________
Abstract : In this paper, a solution for privacy preservation and data security is presented. Privacy over the
internet can be defined as the ability to decide what information one discloses or withholds about a person over
the internet, who can access such informatio n and for what reason a person’s information may or may not be
accessed. The problem is stated as follows: (i) a client needs to inquire a d atabase which contains some
authorized and sensitive data and does not want to disclose himself to the server because of privacy concerns
(ii) the owner of the database i.e the server, does not want to simply give out its data to all users. The server
needs to have some control over its information, since the information is its asset. In this paper, a two stage
approach is proposed to achieve secure solution for both user and the server. The first step is accomplished
using Oblivious Transfer and second step is a ccomplished using Data Retrieval phase. And, a security model
has been devised, which includes encryption and hashing algorithm for providing data security.
Keywords - Privacy preservation, data security, location based services, oblivious transfer, Data Retrieval.
I. INTRODUCTION
Privacy is one of the critical aspect to look at while browsing over the internet since individuals can be
hurt if there are no confinements on public’s access and utilization of individual data. If an unauthorized user
gets access to sensitive personal information of a person like medical records, court recor ds, psychological tests
and interviews, financial records from bank, sites bro wsed over the Internet and a variety of different sources of
information holds numerous close points of interest of a individual’s life. I f such data of an individual are
leaked, it can leave an individual defenseless against a number of misuses. Consider a scenario where, ‘A’ has a
patent database and is not willing to give the entire information present in t he database to various groups, but is
willing to allow individuals or groups to browse the database via ‘World Wide Web’ interface. ‘B’ has a bright
idea which ‘B’ is thinking to patent and so therefore ‘B’ first carries out a search for correlated patents. But, the
concern is about the fact that if ‘B’ directs an inquiry on ‘A’s database ‘A’ might realize what ‘B’ is keen on and
might reveal the idea of ‘B’. Therefore the queries of ‘B’ as to be maintained such that ‘A’ does not know what
‘B’ queries are. Therefore in this particular scenario both p ersonal information and the queries of the user have
to be protected. Under certain conditions, breach in sensitive personal information is so serious that the
individual may be susceptible to blackmail and coercion by those who have admittance to that data. Therefore,
privacy preservation is very important while browsing over the Internet.
Data security can be said as protecting data (Ex: database) from corruption a nd from undesired actions
of unauthorized users. Data security is critical when storing the data in the cloud because if the sensitive data in
the cloud falls into the wrong hands like hackers, it can cause a serious threat to the user and the company
owning the cloud. Ex: If bank account details fall into the hands of a hacker how dangerous can it be. Therefore
for providing data security there are many ways like encrypting the data, authorization, authentication, password
protection, backup of data and by using hashing algorithms.
A location-based service (LBS) is an application for an IP enabled mobile devices that requires the where
about of the location of the mobile device. LBS are query based and it provides information related to the
location of the gadget. Ex: Where are the nearest AT M’s? Or it can be any paid information about the particular
location. Location server provides the LBS to the user. Therefore it is important to secure the privacy of the
user while the user is doing online transactions with a location server and the data in the location server must be
provided security so that it is not accessed by unauthorized users.
II. LITERATURE SURVEY
Jaydep Sen[1] has proposed a safe and proficient scanning scheme for shared systems that uses
topology modification by building up an overlay of trusted peers. Here selection of neighbors is done based on
their trust ratings and content resemblance. Hannes Federrath et al [2] have proposed a devoted DNS Anonymity
Journal of Science and Technology
www.jst.org.in 14 | Page
Service, that ensures privacy preservation of an user. The outline involves two building hinders: a broadcast
plan for conveyance of “top list” of DNS hostnames and low latency mixes for inquiring the hostnames left out
without being viewed. Pericle Perazzo, Gianluca Dini [3] have proposed a method called UNILO, an
obfuscation operator which offers high guarantee on obscurity consistency, even in situations of erroneous
location estimation. Muhammad Aqib and Jonathan Cazalas [4] have proposed caching data technique to resolve
the privacy problems where in, it reduces the quantity of queries requested to t he location server. Jun Shao et al
[5] have p roposed a technique called FINE which is a fine-grained privacy preserving location-based service
(LBS) framework for cell phones. FINE uses data-as-a-service (DaaS) model. Here LBS supplier sends its data
to a third party, who in turn evaluates the users LBS queries. The FINE uses cipher text policy unknown
attribute based encryption method to achieve location privacy, access control, access polic y and confidentiality
of the LBS data and get precise LBS query o utput without concerning any trusted third party. Thomas
Ristenpart et al [6] have proposed a system called Adeona. It provides good surety of location confidentiality
and it has the capability to proficiently find missing devices. Adeona uses OpenDHT as the third party service.
Shahriyar Amini et al [7] have proposed a system named as Cache. This system provides location privac y for
some classes of location-based applications and helpful location enhanced contents are taken in prior.
Applications can get the contents if required from the local cache on the cell phone. This methodology permits
the user to utilize the location enhanced content while simply revealing to third-party content providers what
geographic range she is in instead of her precise area. Femi Olumofin et al [8] have proposed an entr y privacy
method and a framework for questioning huge databases. This strategy investigates offline data classification,
restraint based query transformations and privacy saving questions to record structures much smaller than the
databases. This technique empowers the querying of a large database by statically determining or dynamically
describing database portions on keys, po ssibly with high diversity in the ir collection of values, t hereby
minimizing data leakage about the potential data items of interest to users. Marco Gruteser and Dirk
Grunwald[9] have introduced a middleware design and algorithms that can be utilized by a federal location
broker service. The adap tive algorithms changes the decision of location information along spatial or temporal
magnitude so that it meets the exact anonymity limitations based on the individuals who could be using location
services inside a particular region. Krishna P. N. Puttaswa my and Ben Y. Zhao[10] have told about Location
Based Social Applications (LBSAs). LBS As have adapted to a methodology where in untrusted third party
servers are simply treated has enco ded data stores and functionality of the application will be moved to the
client’s gadget. The area coordinates are encoded, when sharing and can be decoded only by t he client to whom
the data is proposed. M. Bellare and S. Micali [11] have proposed a client and fair pro tocol for safe two-party
communication in the Optimistic model. Here a partially trusted third party is used but however it’s not going to
be involved in standard computation of protocol. Third party is necessary just if there exi sts any interruption in
communication or if one o f the two parties denies or gets out of hand. This protocol guarantees that regardless
of the possibility that one party terminates the protocol at any of the time, the computation is still reasonable for
the other party to communicate via asynchronous network.
Chi-yin chow et al [12] have proposed a framework called Casper has been introduced in which a user
can obtain LBS without him needing to disclose his private location information. Casper involves two main
components- Location anonymiser (trusted third party) and privacy aware query processor. Location anonymiser
withholds the precise region information o f the client into a cloaked region. The privacy aware query processor
is in the location server which tunes the functionality according to cloaked region rather than specific point
information. B. Hoh and M. Gruteser[13] have proposed an algorithm called perturbation algorithm. When two
users path meet, improbability to the location data of the user is added. This will make it difficult to track a user
just using location data.
III. SYSTEM IMPLEMENTATION
Fig : System Architecture
Journal of Science and Technology
www.jst.org.in 15 | Page
The implementation of this paper is as follows: There are four components. They are as follows: 1) End
Users log in via PC or a mobile 2) Intermediate Service Provider (Trusted third party) 3) A location server.
4) Secondary server
An end user first has to register himself with the Intermediate Service Provider (ISP) so that he ca n
have the benefits of the ISP. After registration, the e nd user will be given the login credentials which he has to
use at the time of login. The ISP’s main aim is to protect the individual’s privacy and to protect the data of the
server that is it gets only the data which the user asks for and none of the other data is revealed to the user.
2.1 Location Data upload
The location based d ata will be uploaded into the location server by the Admin of location server.
While uploading the data, it is going to be encrypted and then saved in t he location servers. For each data, a
Message Authentication Code (MAC) will be generated using SHA-1 algorith m. Then a secret key is going to
be generated for the data using RSA algorithm. So therefore, Secret key, MAC and the encrypted data will be
stored in the Primary server and the Secondary server. In secondary server, data is stored for data recovery in
case if an attacker changes or deletes the data in the primary server. In the server, each block of data will be
encrypted using a different secret key.
2.2 Location Data Retrieval by End user
After the end user has registered himself with the ISP, the user can access the location server via World
Wide Web without their personal details or location details being revealed. When the end user wants to retrieve
some information from the ser ver he sends his query regarding the information which he wants to the ISP. ISP
then retrieves the information in two sta ges. They are oblivious transfer phase and Data Retrieval phase (DRP).
In the first phase, Oblivious transfer phase the ISP gets the secret key and the cell ID of the queried information
from the server and gives the end user, the secret key and the cell ID is not disclosed to the end user. In DRP
phase, end user is asked to enter t he secret key obtained and then the data will be retrieved by the ISP a nd
provided to the end user. At the time of data retrieval in the servers, the MACs of the data stored in primary
server and the secondary server are compared. If both the MACs are same then the data in the server is safe and
data will be sent from the primary server to the ISP. And if the MACs are not same then the data in the p rimary
server might be corrupted. So therefore, data from the secondary server will be sent to the ISP and data to the
primary server will be recovered. The end user can decrypt the received data only if he has the right secret key.
2.3 Data Recovery
Consider a scenario, where in the data in the location server is attacked and the contents are modified. The
attacker has modified the contents of the primary server. W hen the end user asks for the data which is attacked
and modified then the following steps take place:
(i) The MACs of the data from the primary server and the secondary server will be compared. Since data in the
primary server is attacked MAC of the two data will not be same.
(ii) So, therefore in this situation data from the secondary server will be sent to the user.
(iii) Then regarding the attac k admin is going to be notified. And then he recovers the data into primary server
from the secondary server.
2.4 Intermediate Service Provider
Intermediate Service Provider is the mediator between the user and location server. Here two phases
namely oblivious transfer and data retrieval phase undergo. The phases work as follows:
2.4.1 Oblivious Transfer Phase
Oblivious Transfer is a protocol where in two parties are involved. At first the sender has a few
information a nd toward the end of the communication the other party, the recipient finds out about this
information in a way where in the sender (server) won't realize what information the recipient learnt. In t his
implementation, oblivious transfer is used to get one and o nly one record from the database. In this paper, K
out of N obli vious transfer protocol is used where in we can retrieve K information from a set of N data
available in the database.
This protocol contains two phases. They are as follows:
2.4,1.1 Initialization Phase
The initialization phase is controlled by the server who possesses the N infor mation components X1, X2….. XN.
Server typically generates an assurance to each of the N data components.