See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/305811409

An Enhanced Password-Username Authentication System Using

Cryptographic Hashing and Recognition Based Graphical Password

Article in IOSR Journal of Computer Engineering · July 2016

DOI: 10.9790/0661-1804015458

CITATIONS READS

2 504

3 authors, including:

Tivkaa Manasseh Luper

Federal University of Agriculture, Makurdi
1 PUBLICATION 2 CITATIONS

SEE PROFILE

All content following this page was uploaded by Tivkaa Manasseh Luper on 04 August 2016.

The user has requested enhancement of the downloaded file.

IOSR Journal of Computer Engineering (IOSR-JCE)
e-ISSN: 2278-0661,p-ISSN: 2278-8727, Volume 18, Issue 4, Ver. I (Jul.-Aug. 2016), PP 54-58
www.iosrjournals.org

An Enhanced Password-Username Authentication System Using

Cryptographic Hashing and Recognition Based Graphical
Password
Tivkaa, M.L.1; Choji, D. N.2; Agaji, I.3; Atsa‟am, D.4
1, 2, 3, 4
(Department of Mathematics/Statistics/Computer Science, University of Agriculture, Makurdi, Nigeria)

Abstract: Password-username authentication is a critical component of today’s web application technology

that is commonly used to control access to restricted resources. However, poor design, coding flaws and weak
user login credentials exposes this functionality to Sequel Query Language Injection (SQLI) and online
password guessing attacks. Current techniques advanced by researchers to address authentication attacks only
focus on either one of them, thus failing to envisage a scenario where the login form can be used to launch both
SQLI and online password guessing attacks. To address this challenge, this paper presents an authentication
solution that addresses the issue of SQLI and online password guessing attacks on login form as implemented in
generic web applications. The solution combines the use of plain text credentials that are cryptographically
hashed at runtime with recognition based graphical login credentials. The goal is to always guarantee access to
a user account even when such account is under attack while at the same time ensuring convenient and secure
login experience by legitimate users. This is achieved by blocking the Internet Protocol (IP) addresses from
which there are unsuccessful login attempts. Security test shows that the solution is not vulnerable to SQLI and
online password guessing attacks.
Keywords: Authentication, Password Guessing, Graphical Password, SQL Injection, Web Application Security

I. Introduction
Web applications have become popular internet services and are critical to the survival of many
enterprises that rely on them. To restrict access to privilege information stored on the web, various
authentication schemes such as biometrics, unregistered user requirements, public key cryptography, keystrokes
dynamics, click pattern, graphical passwords, one-time password, digital signatures, authentication panel, zero-
knowledge proof as well as password/username are implemented in various web applications [1][2]. However,
the most commonly used method of authentication is the password/username combination which is implemented
using traditional HTML login form with input fields that allow users to enter their username and text based
password [1][3].
Though prominent, the vulnerabilities associated with the implementation of Password/Username
authentication pattern in web applications include the threat of password guessing and Sequel Query Language
Injection (SQLI) attacks [4][5]. Various studies have shown that the cause of SQLI is largely poorly sanitized
input from the users [6][7]. Password guessing attacks on the other hand are usually successful against weak
passwords, poor password enforcement policies, and poor design of the authentication functionality as well as
its implementation [8].
To address the problem of SQL Injection Attacks (SQLIAs), researchers have advanced several
techniques ranging from defensive coding best practices to automated frameworks for detection and prevention
of these forms of attacks [9]. In the same way, techniques such as account lock out, Completely Automated
Public Turin Test to Tell Computers and Humans Apart (Captcha) and the use of graphical passwords have
been proposed by researchers to solve the problem of online password guessing attacks on login forms
[8][10][11]. Though these techniques are contextually efficient, they fail short of combining SQLI and password
guessing attacks in one solution, thus researchers have failed to preconceive the possibility of an attacker using
the login form to launch both SQLI and online password guessing attacks.

II. Related Works

2.1 Password Guessing Attacks
A password guessing attack is an attempt by a malicious user to gain unauthorized access to an
application resource by using a repeated set of large word lists to guess login credentials [4]. These attacks
manifest in various forms such as manual, brute-force and dictionary attacks. The cause of password guessing
attacks is the use of weak passwords as well as poor password implementation policies. To defend against this
threat, [12] proposed a system known as the Password Guessing Resistant Protocol. This approach sought to
minimize the use of Completely Automated Public Turing Test to tell Computers and Human Apart (Captcha).

DOI: 10.9790/0661-1804015458 www.iosrjournals.org 54 | Page

 An Enhanced Password-Username Authentication System Using Cryptographic Hashing and….

It enforces Captcha after a limited failed login attempts from a computer that is known to the system or user
which according to the technique is identified by their IP addresses as well as cookies sent by the browser which
are stored as white-list on the server. Though this approach will slow down the rate of attack, it has a drawback
of associating a legitimate user with so many IP addresses that in the long run will be difficult to manage.
Similarly, if the user uses different browsers or more than one operating system on the same machine then it will
be difficult to effectively identify the user in all cases. The identity of the user can also be manipulated if
browser cookies are altered by way of cookie theft.
In a similar way, [13] proposed a hybrid graphical password system that requires the user who has
registered with a username, password and a graphic image to provide the username, password and redraw the
graphic image to match with the stored graphic during the process of authentication. However this system will
be easy to use for the technology savvy users, it will be a challenge for non-technology friendly individuals.
Another drawback of the system will be the challenge of drawing the graphics to exact coordinates as the stored
one. Another issue with this approach is that it is better suited for smart screens making it difficult for users who
do not have such devices to gain authentication.
Another obvious method quite often deployed by developers to mitigate online password guessing
attack is the use of account lock out security pattern [10]. Account lockout simply means denying access to a
specific account after a failed number of incorrect password attempts. Account lockout can last a specific
duration, such as one hour, or the affected account could remain blocked until manually unblocked by the
account administrator. Wrongful implementation of this strategy could lead to a manifestation of other forms of
attacks such as denial of service (DoS). Also if this functionality is not implemented properly it can be a fertile
place for reconnaissance.
2.2 SQL Injection Attacks
SQL Injection is an attack vector that allows a malicious user to inject harmful SQL query statements
that are executed by the application logic [6][11][14]. The cause of SQL Injection attack is when the application
is designed to process user data without proper validation. SQLIAs represent one of the main security issues in
database driven web applications [15]. According to Open Web Application Security Project [16], SQL
Injection attacks remains one of the most dangerous attack vectors today and is on top of their annual report.
SQL injection is mainly launched through web forms such as login functionality and altering query string
parameters. There are various forms of SQLIAs such as tautology, piggy-backed, union queries and blind
injection attacks [17][11][6]. Though their attack methodologies vary, the intent is the same: Data theft, Damage
to corporate image, Denial of Service etc.
To address the challenge of SQLIAs, [18] proposed a technique that checks user credentials against
encrypted values against placeholders in stored procedures. This technique is efficient in attacking code
injection, however the technique is based on the assumption that the developer is approaching the development
in a particular way especially in the coding of stored procedures, if they are written insecurely, then the
approach will be ineffective, again it gives flexibility to the attacker to launch another attack vector such as
password guessing attacks.
Another method that is often applied in solving SQLIAs is the application of defensive coding
techniques such as pattern matching, input validation etc. [19] argue that the best approach in tackling these
vulnerabilities is the use of suitable and applicable defensive and secure coding practices such as proper input
sanitization, use of parameterized queries, guiding against input sources, input type checking and hiding default
application errors. This explains that secure code is rooted in the behavior, orientation and awareness of
developers about security issues as it relates to web applications they develop. This is indeed a convincing
reason for proper usage of defensive coding techniques.
Most of these defensive coding practices come at the expense of performance and ease of use which is
largely due to the fact that security is treated as add-ons rather considering it as part of the goal of the system
right from the design stages [20].
III. Proposed Technique
To mitigate online password guessing and SQLI attacks on our authentication functionality, we present
a technique that combines the use of cryptographic hashing algorithm, recognition based graphical password and
parameterized queries. A randomly generated salt which is stored in the user‟s database table is combined with
the user‟s chosen plain text password to produce and save the final encrypted password at the plain text
registration phase. A user is also expected to setup his graphical login details as a second stage in the
registration process by selecting an image category and choosing a specific image under such category to
complete the registration phase. During authentication a user is allowed two login attempts using his plain text
credentials, the password is computed at runtime using the salt generated at the registration phase. After two
failed attempts, a user is allowed to login using his graphical password. The user who is identified by IP address
is subsequently blocked after one failed graphical login attempt. The entire algorithm is summarized in Fig. 1
below:
DOI: 10.9790/0661-1804015458 www.iosrjournals.org 55 | Page
 An Enhanced Password-Username Authentication System Using Cryptographic Hashing and….

Fig 1: Flowchart of the proposed technique

IV. Materials/Methods
The proposed solution used the Microsoft Secure Development Life-Cycle as the software development
model. The program structure was designed using the .NET Code-Behind programming structure while
ADO.NET 2.0 was used as the underlying data access logic and MS SQL served as the database server.
Microsoft threat modeling methodology was used to model the security of the authentication functionality. To
achieve this effectively, Microsoft Threat Modeling Tool (TMT) was used to draw the application Data Flow
Diagrams (DFDs) which represent different component of the application. The TMT then classified threats
using the in-built Microsoft STRIDE model and rated them using the in-built DREAD model. The proposed
authentication solution has been tested using TamperIE to evaluate the effectiveness of the solution.

V. Results
In this section, we present the result obtained by implementing the proposed technique, which is
deployed locally as a web application over a secured communication channel using Secure Socket Layer (SSL).

Figure 2: SQLI attack scenario

Fig. 2 shows a SQL attack on the login form. Upon submission of a login form, TamperIE
automatically intercepts the data as shown in Fig. 2 without sending it to the server. To manipulate the name
pair values in the TamperIE read/write mode, * „OR „1‟= „1 which is a tautology attack is selected and ran

DOI: 10.9790/0661-1804015458 www.iosrjournals.org 56 | Page

 An Enhanced Password-Username Authentication System Using Cryptographic Hashing and….

against the login form by clicking “send altered data” button at the top right corner. Several other SQLI
constructs in the TamperIE were also selected and ran against the login form. The result was unsuccessful SQLI
automated attacks as seen in Fig. 3.

Figure 3: Failed SQLI attack

Figure 4: Graphical login page attack

Fig. 4 shows TamperIE used to manipulate the graphical login page by changing image Identity (ID)
numbers in an automated attack. This attack scenario is unsuccessful because only one login attempt is allowed
using the graphical login details. It prevents a malicious user from launching effective dictionary or brute force
attacks on a login functionality based on the user logged IP address, this forces the user to request a password
reset while the account is locked against the suspected IP address.
This security feature is important because it does not prevent a legitimate user from accessing his
account even when an account is locked against an IP address. This signifies the failure of Denial of Service
(DoS) through the login functionality. However it makes life difficult for an attacker to launch password
guessing attacks unless the attacker changes IP address which is a cumbersome endeavour in the circumstance.

VI. Conclusion
The paper presents an authentication solution that combines the use of traditional password-username
with graphical image login credentials. The solution gives the user the flexibility and convenience of use while
at the same time maintaining the security considerations regarding the design of the solution. The test result
showed that the use of threat modeling in the design of an application had tremendous impact on the application
security, this is because security considerations were part of the development process.
The result showed that the application was not vulnerable to SQLI and online password guessing
attacks. Remarkably, it was able to block malicious users based on their IP addresses but allowed access to
legitimate users. This technique can be used in web applications that use password-username authentication
scheme. In the future a solution will be proposed to enable the application learn automatically from previous
attacks.

References
[1]. Kiiski, Lauri. "Security Patterns in Web Applications." Publications in Telecommunications Software and Multimedia Laboratory,
Available at: http://www. tml. tkk. fi/Publications/C/25/papers/Kiiski_final. pdf (Last Accessed: November 2011) (2007).
[2]. Jesudoss, A, and N Subramaniam. "A Survey on Authentication Attacks and Countermeasures in a Distributed Environment."
Indian Journal of Computer Science and Engineering, 5(2) (2014).
[3]. Kienzle, Darrell M, and Matthew C Elder. "Final technical report: Security patterns for web application development." DARPA,
Washington DC (2002).
[4]. Dave, Mr Sachin R, and Vaishali B Bhagat. "Defecating online password guessing attack using 3 tier security." International
Journal of Application or Innovation in Engineering & Management 2(12) (2013)

DOI: 10.9790/0661-1804015458 www.iosrjournals.org 57 | Page

 An Enhanced Password-Username Authentication System Using Cryptographic Hashing and….
[5]. Kindy, Diallo Abdoulaye, and Al-Sakib Khan Pathan. "A detailed survey on various aspects of sql injection in web applications:
Vulnerabilities, innovative attacks, and remedies." arXiv preprint arXiv:1203.3324 (2012).
[6]. Gandhi, Mihir. "JwalantBaria, s “SQL Injection Attacks in Web Application” International Journal of Soft Computing and
Engineering 2(6) (2013).
[7]. Balasundaram, Indrani, and E Ramaraj. "An authentication mechanism to prevent SQL injection attacks." International Journal of
Computer Applications, 19 (1) (2011): 30-33.
[8]. Pinkas, Benny, and Tomas Sander. "Securing passwords against dictionary attacks." Proceedings of the 9th ACM conference on
Computer and communications security 18 Nov. 2002: 161-170.
[9]. Halfond, William GJ, and Alessandro Orso. "Detection and prevention of sql injection attacks." Malware Detection (2007): 85-109.
[10]. Scambray, Joel, Mike Shema, and Caleb Sima. Hacking exposed: Web applications. San Francisco: McGraw-Hill, 2006.
[11]. Kumar, K. V. & Das, D. J. “Advanced Detecting and Defensive Coding Techniques to Prevent SQLIAs in Web Applications: A
Survey.” International Journal of Science and Modern Engineering (6) (2013)
[12]. Garg, Nitin, Raghav Kukreja, and Pitambar Sharma. "Revisiting Defences against Large Scale Online Password Guessing Attacks."
International Journal of Scientific and Research Publications 3.4 (2013).
[13]. Khan, Wazir Zada, Mohammed Y Aalsalem, and Yang Xiang. "A graphical password based system for small mobile devices."
arXiv preprint arXiv:1110.3844 (2011).
[14]. Shehu, Bojken, Aleksander Xhuvani, and Shqiponja Ahmetaj. "Methods of Identifying and Preventing SQL Attacks" International
Journal of Computer Science 9(6) (2012).
[15]. Ali, Shaukat, Azhar Rauf, and Huma Javed. "Sqlipa: An authentication mechanism against sql injection." European Journal of
Scientific Research 38(4) (2009): 604-611.
[16]. Wichers, Dave. "OWASP Top-10 2013." OWASP Foundation, February (2013).
[17]. Anley, Chris. "Advanced SQL injection in SQL server applications." 31 Jan. 2002.
[18]. Rani, D.R., Kumar,B.S., Rao, T.R., Jagadish, V.T. and Pradeep, M., (2012) Web Security by Preventing SQL Injection Using
Encryption in Stored Procedure. International Journal of Computer Science and Information Technologies 3(2) (2012):3689-3692
[19]. Borade, Monali R, and Neeta A Deshpande. "Extensive Review of SQLIA„s Detection and Prevention Techniques." International
Journal of Emerging Technology and Advanced Engineering 3(10) (2013).
[20]. Yee, Ka-Ping. "Aligning security and usability." IEEE Security & Privacy 5 (2004): 48-55.

DOI: 10.9790/0661-1804015458 www.iosrjournals.org 58 | Page

View publication stats