Transparent Developmental Biometric Based System Protect User Reauthentication Using Smartphones

International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395 -0056
Volume: 04 Issue: 3 | Mar -2017 www.irjet.net p-ISSN: 2395-0072
© 2017, IRJET | Impact Factor value: 5.181 | ISO 9001:2008 Certified Journal | Page 591
TRANSPARENT DEVELOPMENTAL BIOMETRIC BASED SYSTEM
PROTECT USER REAUTHENTICATION USING SMARTPHONES
*1 Mr. R.Arunkumar, *2 M.Arunraj, *3 R.Prabhakaran, *4 Mr.V.Naveen
*1,2,3 B.Tech Student, Department of Information Technology Kingston Engineering College, Christainpettai,
Katpadi, Vellore.
*4 Assistant Professor, Department of Information Technology Kingston Engineering College, Christainpettai,
Katpadi, Vellore.
---------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - With the emergence of smartphones as an
essential part of daily life, the demand for user
reauthentication has increased manifolds. The effective
and widely practiced biometric schemes are based upon
the principle of “who you are” which utilizes inherent
and unique characteristics of the user. In this context,
the behavioral biometrics such as sliding dynamics
make use of on-screen sliding movements to infer the
user’s patterns. In this paper, we present Safeguard, an
accurate and efficient smartphone user
reauthentication (verification) system based upon on-
screen finger movements.. The key feature of the
proposed system lies in fine-grained on-screen
biometric metrics, i.e., sliding dynamics and pressure
intensity, which are unique to each user under diverse
scenarios. We first implement our scheme through five
machine learning approaches and finally select the
support vector machine (SVM)-based approach due to
its high accuracy. We further analyze Safeguard to be
robust against adversary limitation. We validate the
efficacy of our approach through implementation on
off-the-shelf smartphone followed by practical
evaluation under different scenarios.
Key Words: SVM, honeybeeandantcolony, FAR,FRR,
reauthentication with smart phone, Mobile
Verification.
I. INTRODUCTION
Mobile Devices, especially smartphone, becomes
the most common used tools for human daily accessing
mobile social networks (MSNs) [1]-[3]. Thus, preserving
users' privacy and sensitive data grows to the urgent need
as smartphones store users' more private and sensitive
information. At present, many smartphone authentication
schemes are With the introduction of the Android
operating system for mobile phones, an alternative to PIN-
authentication on mo-bile devices was introduced and
widely deployed for the first time. The password pattern,
similar to shape-based authentication approaches. Despite
its manifold advantages, this approach has major
drawbacks, the most important one being security. Drawn
passwords are very easy to spy on [11,36], which makes
shoulder surfing, a common attack in public settings [27],
a serious threat. Other attacks include the infamous
smudge attack [1], in which finger traces left on the screen
are used to extract the password. Due to its weak security
properties, this authentication approach does not fully
meet the requirement of adequately protecting the user’s
data stored on the device. Nowadays, not only private but
also valuable business information is stored on the user’s
handheld [10]. Therefore, resistance to attacks is a major
concern when designing respective authentication
systems.
The biometric-based authentication has been widely
used in many applications such as access control and
continuous tracking systems [5]. However, the
reauthentication is still challenging on smartphones. The
reason lies in three-folds: 1) Current biometric
approaches, such as fingerprints and retinal [6] scans,
provide accurate one-time authentication but require spe-
cialized hardware, which may be nontransparent,
expensive, or unavailable for smartphones. 2) A typical
smartphone is equipped with a touch screen as the user
interface. Thus, the classical behavioral biometrics on PCs,
such as keystroke and mouse dynamics, cannot be applied
on smartphones for user verification. 3) Existing
reauthentication schemes for smart-phone are not
practical for real applications because of low accuracy [7],
limited application scenarios [8] or high system overhead
[10]. Moreover, most existing approaches only address the
one-time authentication when users unlock the screen of
the smartphones.
.

We employ machine learning to devise an efficient
system and use support vector machine (SVM) for optimal
results. More in detail, the sliding movements of a
legitimate user are mapped from touch screen in a passive
and transparent way. Same information is also used to
extract the angle metric, which is a unique feature of
user's sliding movements. In parallel, the pressure-based
metrics are calculated as the user applies his/her fingers
on the screen. With these features, Safeguard "trains" a
model of the user's behavioral biometrics consisting of on-
screen pressure and the sliding movements (curve, angle,
distance ratio). After the training, Safeguard records the
new biometric parameters and compares them with the
modeled profile of the user. After the classification, a
decision is made based upon the similarity threshold on
whether new parameters relate to a valid user or not. In
essence, our key approach is to exploit the behavioral
features by means of angle and pressure-based metrics
which are the outputs of user's sliding movements and
intensity of pressure on the touch screen. Both of these
metrics are observed to be unique for each user and
distinct from person to person.
Silent Sense is designed as a pure software-based
framework, running in the background of smartphone,
which nonintrusive explores the behavior of users
interacting with the device without any additional
assistant hardware. The main idea of Silent Sense for user
identification comes from two aspects: (1) how you use
the device; and (2) how the device reacts to the user
action. While using mobile devices, most people may
follow certain individual habits unconsciously. Running as
a background service, Silent Sense exploits the user’s app
usage and interacting behavior with each app, and uses
the motion sensors to measure the device’s reaction.
Correlating the user action and its corresponding device
reaction, Silent Sense establishes a unique biometric
model to identify the role of current user.
II. RELATED WORK
The underlying principle of Biometric-based user
authentication focuses on "who you are" which differs
from conventional user authentication approach that
mainly relies on "what you have" or "what you know."
Thus, a biometric-based approach is based on the inherent
and unique characteristics of a human user being
authenticated. Biometric-based reauthentication
approaches have been widely studied for PCs [8]-[7].
However, we can find only few such implementations on
smartphones, which are either limited by coarse accuracy
or restrict application scenarios or gestures. Therefore, in
this section, we focus on the state-of-the-art on
smartphones. We first present some smartphone
applications that perform the one-time identification and
reauthentication.
User Behavior Modeling: To characterize individuals’
unconscious use habits accurately, the user behavior
model should contain multiple features of both user’s
action and device’s reaction. In addition, the connection
between features may not be neglected for identification,
such as different interaction coordinates on the touch
screen may cause different reaction vibrations of the
device.
Identification Strategy: To establish the user behavior
model, for the owner there are abundant behavior
information. For a guest, the collected behavior
information may be very limited. In addition, in motion
scenarios, some interacting features will be swamped by
the motion from the perspective of sensory data, which
greatly increases the difficulty of accurate identification.
Thus it is challenging to distinguish users with limited
information effectively, even if the interacting features are
partially swamped.
Balance Among Accuracy, Delay and Energy: Nonstop
observation with sensors provides identification with
small delay and high accuracy, but may cause unwanted
energy consumption for the mobile device when the
current user is the owner or a guest is using an insensitive
app, e.g., playing a game. But intrusion may happen when
the sensors are off and the risk increases with the
detection delay. A well designed mechanism is required to
decide the observation timing to reduce energy
consumption while guarantee the identification accuracy
and delay.
Fig : System Framework
The framework model consists of two basic phases:
Training and Identification, as shown in Figure 1. The
training phase is conducted to build a behavior model
when the user is interacting with the device, and the
identification phase is implemented to distinguish the
identity of the current user based on the observations of
each individual’s interacting behaviors. When a guest user
is observed, privacy protection mechanism will be
triggered automatically. After the guest leaves and the
owner returns, privacy protection will be reset for the
owner’s convenience
Initially, we assume that the device has only the owner’s
information, e.g. a newly bought phone. The framework
trains the owner’s behavior model by retrieving two types

of correlated information, the information of each touch
screen action and the corresponding reaction of the device
when the user is static. In the motion scenario, instead of
the reaction information, motion features will be detected.
With the owner’s behavior model, the current user will be
identified through one-class SVM classification. One-class
SVM provides a judgment whether the observed features
belong to the owner (true of false). However the
identification accuracy by one observation is usually not
high enough for an identity conclusion, continuous
consistent judgments will increase the accumulated
confidence for this judgement.When the accumulated
confidence is high enough, a conclusion is ready and the
newly observed features will be added to the owner or
guest dataset according to the conclusion to update the
model.
Initially, without only the owner’s behavior data, a one-
class SVM model is trained to identify a new observation Oi
and provide the judgment Ji whether this action belongs to
the owner or not, i.e., Ji = true or Ji = false. Lacking of
ground truth, it is difficult to determine the correctness of
the judgment. To achieve high identification accuracy, we
adopt the SVM model’s credibility for each judgment Ji as
the confidence of the framework on Ji. Let this confidence
be "i(Ji), which indicates the probability the framework
considers that Ji is correct for the current observation Oi.
Using one-class SVM, the judgment of one observation
usually is not accurate enough to make a identity
conclusion. Obviously, more observations leads to higher
conclusion accuracy. Specifically, let {J1, J2, · · · , Jk} be a
sequence of consistent judgments, i.e. J1 = J2 = · · · = Ji, to
continuous observations {O1,O2, · · · ,Ok}. Based on the
judgment sequence, an identity conclusion I1,k can be
made, i.e. I1,k = Jk. Then the conclusion confidence will be
cumulated as
P1,k = P(J1, J2, · · · , Jk) = (1- π (1 − "i(Ji))) –(1).
Which indicates the probability this framework
considers that the identity conclusion I1,k is correct. Then
the identification delay dk for a conclusion I1,k is defined as
the number of observations taken to achieve this
conclusion. With the number of observation increases, the
framework will be more confident to provide a correct
conclusion, meanwhile the delay will increase. Note that,
an inconsistent judgment will interrupt the sequence, and
the conclusion confidence needs to be cumulated from
scratch.
III. PREVIOUS IMPLEMENTATIONS
At present, many smartphone authentication
schemes are. in practice including PIN and sliding pattern
unlock which are the most common. These mechanisms,
however, give one-time security and cannot protect users’
privacy in run-time. For example, an adversary can
operate a user’s phone once the PINis stolen or
compromised, and can easily abuse the owner’s Private
information. In addition, the verified users can still fall
victim to both session hijacking and leakage of the secret
information. Thus, to guarantee the user authenticity,
more frequent user verification is needed. However,
currently practiced user verification and reauthentication
methods are not suitable for frequent verification, as they
are not passive or transparent to users. Like, frequently
answering predefined secure questions is too obtrusive
and inconvenient to be acceptable. This necessitates a
user-re authentication scheme which should be
transparent and easily implementable in run-time without
any user involvement and system overhead.
Existing smartphones are equipped with sensors such as
GPS, microphone, accelerometer, magnetic field, gravity,
temperature, and gyroscope. These sensors provide a side
channel to compromise the privacy of smartphone owners
as they can record some biometric characteristics [23]. By
accessing the camera and microphone of a smartphone, its
location can be successfully identified [2] Cai and Chen [6]
presented an effective method to infer the sliding unlock
pattern by observing finger smudges on a smart phone’s
screen. An interesting attack called Tap prints in [5]
exploits the gyroscope readings and guesses the on-screen
positions. The basic idea is to use the gyroscope to record
vibrations caused by user pressing on different positions
of the screen. Through machine learning, this attack can
effectively recognize user typing on the virtual keyboard
of a smartphone with the equal error rate (EER) not less
than 30%.
Drawback in previous implementation Users tend to
choose short passwords or Passwords that square
measure straightforward to recollect. Sadly, these
passwords is simply guessed or broken. Another
technique is statistics, like fingerprints, iris scan or
identity verification has been introduced however not
nonetheless wide adopted. square measure graphical
positive. Identification schemes that are planned that
square measure proof against shoulder-surfing, however
they need their own drawbacks like usability problems or
taking longer for users to login.
One-Time Identification: This method enhances the
security of sliding unlock pattern on smartphones by
employing the intensity of pressure on touch screen [13].
Shahzad et al. [14] present an approach for unlocking
smartphone by gestures. Zheng et al. [12] also present an
approach for enhancing PIN with behavior biology.
Basically, these approaches first build models for different
behaviors on touch screen, and then make use of
classification algorithms for user identification. However,
these approaches can provide enhanced security only for
one time while the user is unlocking the smartphone. They
do not provide continuous authentication during user
operations which makes them a nontransparent process.
Moreover, each model for a particular sliding movement
on touch screen, say sliding up, needs to be trained

separately which incurs extra training overheads and
restricts the application to limited behaviors. Vu et al. [28]
develop a special hardware token device which makes use
of specific charging and discharging of capacitors
IV. SYSTEM IMPLEMETNATION
Proposed the transparent biometric-based
reauthentication Approach, called Safeguard, which is
used to verify the users based upon their passive
observable behaviors, i.e., the sliding curves made by the
user as smartphone is used and pressure intensity applied
on the touch screen. Both, sliding movements and touch
screen pressure intensity are unique and Distinguishable
parameters of each user and form the core of our design
methodology due to two reasons. the First security check
we match the angle of the mobile. In this, firstly user has to
match the correct angle for unlocking the mobile that is X,
Y, Z co-ordination. And this is calculated on the base of
rotation of mobile. The option is available to choose the
number of angles is included in the password. The user
wants to select 1 angle, 2 angle or all the 3 angles it
depends on the user. When all the three angles are
selected, then security will be very high. In the particular
mobile rotation the corresponding change in X, Y and Z
value. And in the second step two security checks are done
at one time. In second security check we use the password
pattern. The Password pattern is like draw-a secret shape
on the screen. a user unlocks the mobile first time, we
match time taken to draw with initial time (when user
changes the pattern password), and it store his/her
average time of initial and first time taken to draw
patterns.
Fig : System Model
Advantages of Biometric identification can
provide extremely accurate, secured access to
information; fingerprints, retinal and iris scans produce
absolutely unique data sets when done properly. Current
methods like password verification have many problems
(people write them down, they forget them, they make up
easy-to-hack passwords). Automated biometric
identification can be done very rapidly and uniformly,
with a minimum of training. Your identity can be
verified without resort to documents that may be stolen,
lost or altered.
SYSTEM ARCHITECTURE
Fig : System Architecture
Key Contributions
In this paper, we make following five key contributions.
 We proposed, implemented, and evaluated a
gesture based authentication scheme for the
secure unlocking of touch screen devices.
 We identified a set of effective features that
capture the behavioral information of performing
gestures on touch screens.
 We proposed an algorithm that automatically
segments each stroke into sub-strokes of different
time duration where for each sub-stroke the user
has consistent and distinguishing behavior.
 We proposed an algorithm to extract multiple
behaviors from the training samples of a given
gesture.
 We collected a comprehensive data set containing
15009 training samples from 50 users and
evaluated the performance of GEAT on this data
set.
Algorithm1: Honey Bee and Ant Colony Algorithm
Step1.Initialise population with random solutions.
Step2. Evaluate fitness of the population.
Step3. While (stopping criterion not met)
//Forming new population.
Step4. Select sites for neighbourhood search.

Step5. Recruit bees for selected sites (more bees for best e
sites) and evaluate fitnesses.
Step6. Select the fittest bee from each patch.
Step7. Assign remaining bees to search randomly and
evaluate their fitnesses.
Step8. End While.
Algorithm Implementation
For all i such that i is a valid path number do
If obpa (i) < cwnd(i) then
Transmit on path i until obpa (i) = cwnd(i);
If more data is pending transmission in queue then
Choose next path i;
End
End
End
If no path was available for transmission and data is
pending transmission
Then
Find path j that has the minimum ratio of obpa (j)
Cwnd (j) over all paths;
Transmit one MTU of data on path j;
End
Algorithm 2: SVM Supported Vector Machine
Data: PathSet ← set of valid path numbers sorted in
ascending order of the number of packets missing on the
respective paths
Suppose we choose the Threshold (seen below) that is
close to some sample Now suppose it have a new point
that should be in class “-1” and is close to . Using our
classification function his point is misclassified! -The SVM
idea is to maximize the distance between The hyperplane
and the closest sample point.
Let j be the first path from PathSet;
While data is pending transmission do
If (obpa(j) < cwnd(j)) or (missing(j) + sentAlready(j) <
missing(j +1))
Then
transmit on path j;
End
Else
Choose path j + 1;
If j + 1 is not a valid path number then
Break out of the loop;
End
End
A) Smartphone Sensors :
Most modern smart phones have built-in sensors which
can measure motion and environmental and positional
environment the devices are subject to. They provide
several facilities such as providing accurate and precise
raw data, observing the position in three dimensions, and
measuring any possible changes in the surrounding
environment sufficiently close to the device.
 Motion sensors These sensors measure
acceleration forces and rotational forces along
three axes. This category includes accelerometers,
gravity sensors, gyroscopes, and rotational vector
sensors.
 Environmental sensors These sensors measure
various environmental parameters, such as
ambient air temperature and pressure,
illumination, and humidity. This category includes
barometers, photometers, and thermometers.
 Position sensors These sensors measure the
physical position of a device. This category
includes orientation sensors and magnetometers.
Sensor Coordinate System
In general, the sensor framework uses a standard
3-axis coordinate system to express data values. For most
sensors, the coordinate system is defined relative to the
device's screen when the device is held in its default
orientation (see figure 1). When a device is held in its
default orientation, the X axis is horizontal and points to
the right, the Y axis is vertical and points up, and the Z axis
points toward the outside of the screen face. In this
system, coordinates behind the screen have negative Z
values. This coordinate system is used by the following
sensors:
Coordinate system (relative to a device) that's used by the
Sensor API.

 Acceleration sensor
 Gravity sensor
 Gyroscope
 Linear acceleration sensor
 Geomagnetic field sensor
Fig: Sensor Detection
The most important point to understand about this
coordinate system is that the axes are not swapped when
the device's screen orientation changes—that is, the
sensor's coordinate system never changes as the device
moves. This behavior is the same as the behavior of the
OpenGL coordinate system. Another point to understand is
that your application must not assume that a device's
natural (default) orientation is portrait. The natural
orientation for many tablet devices is landscape. And the
sensor coordinate system is always based on the natural
orientation of a device.
Fig : Finger Print Sensor
. Image Authentication:
In this type the authentication user select a points
from the given image. In the time of login the application
allow in the system when the selected region and
previously selected regions are match. The Fingerprint
Manager (and its Support Library counterpart, Fingerprint
Manager Compact) is the primary class for using the
fingerprint scanning hardware. This class is an Android
SDK wrapper around the system level service that
manages interactions with the hardware itself. It is
responsible for starting the fingerprint scanner and for
responding to feedback from the scanner. This class has a
fairly straightforward interface with only three members:
 Authenticate – This method will initialize the
hardware scanner and start the service in the
background, waiting for the user to scan their
fingerprint.
 Enrolled Fingerprints – This property will
return true if the user has registered one or more
fingerprints with the device.
 Hardware Detected – This property is used to
determine if the device supports fingerprint
scanning.
Fig : Image Authentication
Draw line Authentication:
In this type the authentication user draw a line by
connecting two points from the given image. For this
system will take the starting and ending point for
Authentication. During login only allow in to the system
when both the lines are equal.
1. Data Structuring and Processing
Sliding metrics: a raw sliding movement can be
represented as a series of successive points. From here, we
can extract the tuples of Cartesian coordinate pairs,
pressure intensity, size, and timestamp at each point. We
record the biometric values after each time interval of
roughly 17 ms, which serves as a single sample point
1) Direction: For any two consecutive points A and
B,we record the direction along the line AB from
the first point to the second. The direction is
defined at the angle a between the line AB and
the horizontal axis.
2) Angle: For any three consecutive points A , B,
and C ,the angle of the curvature is Z A BC, i.e., the
angle between lines ABB and Bs C.
3) Distance ratio: For any three consecutive points
A , B, and C, consider the length of line AC. The
distance ratio is the ratio of the length of the

perpendicular distance from midpoint B to AC to
AC .
4) Sliding duration: The time duration between a
finger touches and leaves the screen.
4) Sliding distance: The Euclidean distance between
the locations where a finger touches and leaves
the screen.
5) Sliding velocity: The ratio of sliding duration and
sliding distance.
2. Data Feature Exploration
At first glance, it seems that we should use all available
features to identify users. However, including too many
features can worsen performance in practice because of
their varying accuracies and potentially conflicting
signatures.
3. Preprocessing
1 ) Data Grouping and Filtering: The sliding
movements of a single user may have multiple shapes. In
addition, the effects such as screen vibration and finger
joggling bring outliers and incur deviations from the ideal
slide curvature. In Safeguard, we employ data processing
to classify the valid sets of sliding curvatures and to filer
the redundant samples. As a result, the filtered samples,
called the effective samples, are used for classification in a
more precise and efficient way.
Fig : Draw line Authentication
CONCLUSION
In order to prevent unauthorized usage, we have
proposed a re-authentication system using user finger
movement. The system performs continuous
reauthentication and does not need human assistance
during re-authentication. We carried out all our data
collections and experiments on Motorola Droid phones,
which are equipped with low-end touch screens. Therefore,
some metrics may be dropped due to the smartphone’s
hardware limitations. For example, we left out the pressure
related metrics because the touch screen did not provide
accurate pressure measurements. The metrics may need to
be carefully tested or even re-designed before deploying our
system on another smartphone platform. For example,
pressure may become useful and provide more user
information on some smartphone platforms. However, our
work provides a guideline for the metric design on other
platforms and our methodology can still be adopted. Our
work shows that using gestures to construct an
unobservable continuous re-authentication on smartphones
is practical and promising. We have discussed biometric
feature design and selection for finger movement. We have
demonstrated the effectiveness and efficiency of our system
in extensive experiments.
REFERENCES:
[1] M. Dong et al., "Quality-of-experience (qoe) in
emerging mobile social networks," IEICE Trans. Inf. Syst.,
vol. 97, no. 10, pp. 2606-2612, 2014.
[2] M. Dong et al., "Qoe-ensured price competition model
for emerging mobile networks," IEEE Wireless Commun.,
vol. 22, no. 4, pp. 50-57, Aug. 2015.
[3] K. Wei et al., "Camf: Context-aware message
forwarding in selfish mobile social networks," IEEE Trans.
Parallel Distrib. Syst., vol. 26, no. 8, pp. 2178-2187, Aug.
2014.
[4] Y. Zhang et al., "The security of modern password
expiration: An algorithmic framework and empirical
analysis," in Proc. 17th ACM Conf. Comput. Commun. Secur.
(CCS), 2010, pp. 176-186.
[5] A. A. E. Ahmed and I. Traore, "Anomaly intrusion
detection based on biometrics," in Proc. IEEE Inf. Assur.
Workshop, 2005, pp. 425-453.
[6] C. W. Shanley et al., "Remote retinal scan identifier,"
U.S. Patent 5,359,669, Oct. 1994.
[7] T. Feng et al., "Tips: Context-aware implicit user
identification using touch screen in uncontrolled
environments," in Proc. 15th ACM Workshop Mobile
Comput. Syst. Appl. , 2014, p. 9.
[8] P. Saravanan et al., "Latentgesture: Active user
authentication through background touch analysis," in
Proc. 2nd ACM Int. Symp. Chin. CHI, 2014, pp. 110-113.
[9] M. Frank et al., "Touchalytics: On the applicability of
touchscreen input as a behavioral biometric for
continuous authentication," IEEE Trans. Inf. Forensics
Security, vol. 8, no. 1, pp. 136-148, Dec. 2012.
[10] C. Bo et al., "Silentsense: Silent user identification via
touch and movement behavioral biometrics," in Proc. 19th
ACM Annu. Int. Conf. Mobile Comput. Netw. (Mobicom),
2013, pp. 187-190.
[11] L. Li et al., "Unobservable re-authentication for
smartphones," in Proc. 20th Annu. Netw. Distrib. Syst. Secur.
Symp. (NDSS), San Diego, CA, USA, Feb. 24-27, 2013.

[12] N. Zheng et al., "You are how you touch: User
verification on smart-phones via tapping behaviors," in
Proc. IEEE 22nd Int. Conf. Netw. Protoc. (ICNP), 2014, pp.
221-232.
[13] A. De Luca etal., "Touch me once and i know it's you!:
Implicit authentication based on touch screen patterns," in
Proc. SIGCHI Conf. Human Factors Comput. Syst. , 2012, pp.
987-996.
[14] M. Shahzad et al., "Secure unlocking of mobile touch
screen devices by simple gestures: You can see it but you
can not do it," in Proc. 19th ACM Annu. Int. Conf. Mobile
Comput. Netw. (Mobicom), 2013, pp. 39-50.
[15] T. Buch et al., "An enhanced keystroke biometric
system and associated studies," in Proc. CSIS Res. Day,
New York, NY, USA, 2008.
[16] F. Monrose and A. D. Rubin, "Keystroke dynamics as a
biometric for authentication," Future Gener. Comput. Syst.,
vol. 16, no. 4, pp. 351-359, 2000.
[17] F. Bergadano et al., "User authentication through
keystroke dynamics," ACM Trans. Inf. Syst. Secur., vol. 5,
no. 4, pp. 367-397, 2002.
[18] F. Monrose et al., "Password hardening based on
keystroke dynamics," Int. J. Inf. Secur., vol. 1, no. 2, pp. 69-
83, 2002. [19] H. Saevanee and P. Bhattarakosol,
"Authenticating user using keystroke dynamics and finger
pressure," in Proc. 6th IEEE Consum. Commun. Netw.
Conf., 2009, pp. 1-2.
[20] A. A. E. Ahmed and I. Traore, "A new biometric
technology based on mouse dynamics," IEEE Trans.
Dependable Secure Comput.,vol.4, no. 3, pp. 165-179,
Jul./Sep. 2007.
[21] N. Zheng et al., "An efficient user verification system
via mouse move¬ments," in Proc. 18th ACM Conf. Comput.
Commun. Secur. (CCS), 2011,

Transparent Developmental Biometric Based System Protect User Reauthentication Using Smartphones

More Related Content

What's hot (18)

Similar to Transparent Developmental Biometric Based System Protect User Reauthentication Using Smartphones (20)

More from IRJET Journal (20)

Recently uploaded (20)

Transparent Developmental Biometric Based System Protect User Reauthentication Using Smartphones