END USER AUTHENTICATION (EUA)
George S. Oreku, Jianzhong Li
Department of Computer Science and Engineering HIT, P.O.Box 773, 92 Xi Dazhi Street
Nangang District, Harbin 150001 China
Fredrick J. Mtenzi
School of Computing, Dublin Institute of Technology, Dublin 8, Ireland
Keywords: Authentication, Tickets, End-User, Authorization, Kerberos.
Abstract: Authentication is one among a set of services that constitute a security sub-system in a modern computing
or communications infrastructure. End User Authentication flexibility model would proposed in this paper
allow the user to have multiple authentication mechanisms with varying levels of guarantee, and for
suppliers to request and rely on mechanisms appropriate to the service requested. Authentications to end-
user in a simple three level ticket request model algorithms solution on open distributed environment. This
paper describes the ticket used by clients, servers, and Kerberos to achieve authentication toward prevention
of unauthorized access to in sourced data on applications level. However we explore an approach to end
user authentication that generalizes the notion of a textual password that, in many cases, improves the
security. Our approach is based on the use of Kerberos authentication technique and Diffie-Hellman Key
exchange.
1 INTRODUCTION
For the vast majority of computer systems,
authentication for users and passwords are the
method of choice for access control security
mechanism. In consumer applications as diverse as
financial transactions, remote computer login,
building access control, and keyless entry is
extremely important for prove. With that idea in
mind in distributed environment, three approaches to
security can be envisioned:
1. Rely on each individual client workstations to
assure the identity of its user or users and rely
on each server to enforce a security policy based
on user based identifications (ID).
2. Require that client system authenticate
themselves to server, but trust the client system
concerning the identity of its user.
3. Require the user to prove identity for each
service invoked. Also require that servers prove
their identity to client.
Raising questions to eCommerce security this
paper presents authentications and authorization
service model algorithms to an end user by the use
of textual password. We extend the use of the last
models by Diffie-Hellman Key Exchange A Non-
Mathematician’s Explanation (Palmgren,2005) and
Kerberos authentication model (Steiner et al,1998)
quoting at length to place authentication in proper
systems context use.
2 EUA MODEL ARCHITECTURE
Internet
Policy exchange
Appl et
Certif icates
URL
Application to ISP
Authentication
Service Exchange
DBMS
ACTIVE CONTENT
HTML
JAVA
SCRIPTS
ACTIVE CONTROLS
Executable
Java Applets
Behaviour inspection
Behavi our
Profile
Authentication
ticket for access
Level 1
Level 2
2nd time
1st time
2nd time
Allow
Level 3
File Access
Registry Access
Network Access
Browser Setting
changes
Operating Systems
Unit 1
Unit 2
Unit 3
Figure 1: Architecture of a EUA access model.
406
S. Oreku G., Li J. and J. Mtenzi F. (2007).
END USER AUTHENTICATION (EUA).
In Proceedings of the Third International Conference on Web Information Systems and Technologies - Internet Technology, pages 406-409
DOI: 10.5220/0001278604060409
Copyright
c
SciTePress
Our model is built on the premise that defensive,
or exclusionary, security must be aligned with
inclusionary tools and practices that allow users to
access systems and information anytime, anywhere.
We illustrate the architecture of the access for an
end User Authentication as follows:
1 The “early request ticket for the key exchange
authentication from the DBMS unit” that
provides identification that lives beyond the
span of single first level authorizations or
interaction
2 Level two makes it possible to associate
successive message or request to third level.
3 The single use authorization, which conceals
user’s identities and also limits linkage among
given users successive actions is in third level.
4 The three different units (Unit 1, Unit 2 and
Unit 3) verify the user in application-level by
sending different request in each unit.
2.1 Message Exchange Algorithm
We analyze the semantics basing on Kerberos
authentication technique key exchange (Palmgren,
2005) and Diffie-Hellman Key exchange of each
security transcoding by using an online contracting
scenario.
Basic notations
C = Client, ADBMS = Authentication database
Management System, WS = Web Source, IDc =
Identifier of user on C, Idws= Identifier of Web
source, P
c
= Password of user on C, Kws
= secret
encryption key shared by ADBMS and WS, TS =
timestamp, || = concatenation
Steps
(6) With the ticket, C can now apply to (WS) for
service by sending a message to (WS)
containing C’s ID and the ticket.
(6.1) (Ws) decrypts the ticket and verifies that
the user ID in the ticket is the same as the
unencrypted user ID in the message.
(6.2) If the two matches, the server considers
the user authenticate and grant the
requested service.
(7) Simply stated:
(6.1)
C Ú ADB : IDc || Pc || IDws
(6.2)
ADBMS Ú
C: Ticket
(6.3)
C Ú WS: IDc || Ticket
Authentication Service Exhange: To obtain Ticket-
Granting Ticket.
(1)
C Ú ADBMS: IDc || IDtgs ||TS1
(2) ADBMS Ú
C:EKc [Kc,tgs|| IDtgs || TS2 || Lifetime2 || Tickettgs]
Ticket-Granting Service Exchange: To obtain
Service-Granting Ticket.
(3)
C Ú TGS: IDv ||Tickettgs ||Authenticator C
(4) TGS Ú
C: EKc [Kc,
¨ws
|| IDws || TS4 || Ticketv]
Client/WebService Authentication Exhange: To
Obtain Service
(5)
C Ú WS: Ticketv || Authenticator C
(6) WS Ú C: EKc,v[TS5 +1]
3 EUA ASSESSMENT
Considering today’s pervasiveness of malicious
software (Viruses, Trojan horses) and phishing
attacks, any authentications solution must be
resistant against offline credential stealing attacks.
For this we propose a Challenge/response-based
one-time passwords authentication.
3.1 Passwords
We represent these requirements (Jermyn et al 1999)
in an authentication system consisting of five
components. Components are defined inductively as
follow:
1.
The set A of authentication Information is the set
of specific information with which entities prove
their identities.
2.
The set C of complementary Information is the set
of information that the system stores and uses to
validate the authentication information.
3.
The set F of complementation functions that
generate the complimentary from the
authentication information that is: -
For
F, F : A C
f
The set L of
authentication functions that verify
identity. That is for
L, : A × C {true, false}∈→
The set S of
selection functions that enable an entity
to create or alter the authentication and
complementary information.
The goal is to find an
a
A such that, for f
F,
f (a) =
c
c and c is associated with a particular
entity (or any entity).Because one can determine
weather
a is associated with and entity only by
computing
f (a) or by authenticating via | (a) we
have two approaches for protecting the password,
used simultaneously
1.
Hide enough information so that one of a, c, or f
can not be found.
2. Prevent access to the authentication functions L
END USER AUTHENTICATION (EUA)
407
3.2 Security Considerations
In both approaches, the goal of the defenders is to
maximize the time needed to guess the password.
- Let
P be the probability that an attacker guess a
password in specified period of time
- Let
G be the number of guesses that can be tested
in one time unit
- Let
T be the number of time units during which
guessing occurs
- Let
N be the number of possible passwords
Then
TG
N
P
(1)
Example1: Lets password be composed of
characters drawn from an alphabet of 96
characters. Assume that 10
4
guesses can be tested
each second. We wish the probability of a success
guess to be 0.5 once a 365-days period. What is the
minimum password length that will give us this
probability?
From the formula above, we want
()
11
4
365 24 60 60 10
TG
N6.3110
P
0.5
×××
≥= =×
(2)
Thus we must choose an integer s such that
s
i
96 N
i0
=
This holds when
s
6 so, to meet the desired
conditions password of at least length
6 must be
required.
Assumptions
underlie example:
1. The time required to test a password is constant
2. All passwords are equally likely to be selected
Proof:
(a) The first is reasonable, because the algorithms
used to validate password are fixed and either
the algorithm are independent of the password’s
length or the variation is negligible.
(b) The second assumption is a function of the
password selection mechanism. (Morris and
Thomson, 1979)
3.3 One – time Passwords
One time password is a password that is invalidated
as soon as it is used. It uses techniques first
suggested to generate the password. (Lamport, 1981)
With this technology our System takes the
“seed” user enters and generates a list of
n
passwords. The implementation presents each
password as a sequence of
six shorts words (but the
internal presentation is an Integer).
1. User supplies his name to the server
2. The server replies with the number
i stored in the
ticket file
3. User supplies the corresponding password
pi
4. The server computes
(
)
()
hp =hk =k =p
ni+1
i
ni+2
il
and compares the results with the stored password.
If they match, the ticket file to
i1 and stores Pi in
the file. If the authentication fails the ticket file is
left unchanged
Note:
h – one-way hash function, K – Seed.
3.4 Performance Analysis
We will derive and analyze the robustness of
passwords capabilities by probability letting
unauthorized end-user guessing capability to allow
its login within a particular time.
Parameters:
X: password length, Z: time.
The probability that randomly guessed capability
will pass a particular authorization is given by
()
X
1
PX,Z =1-1-
Z
2
⎛⎞
⎜⎟
⎜⎟
⎝⎠
(3)
and the probability that a randomly guessed
capability will pass all d authorization in a system is
simply
(
)
d
zxp , .
Recall that the capability password authorization
must be > than 6 symbols (characters). Figure 2 and
3 shows login performance over time and different
password length.
We leave the exact timing decisions to
s6
symbols, and simply assume in our experiments that
x is likely to be from two to six. From the graph the
probability log in comes close to matches as the
numbers of input values increases within different
time. The performance is excellently, 97.14% of the
attack using short key length per log in
from
(
)
z=1 , and increasing probability to 100% of
login match to key length six with a marking of
running time
(
)
4z . As we show in figure 3, X
must be at least
2, because a valid capability is 100%
that matches any of the
X capabilities in the DBMS,
small value of
X provide the smallest probability that
a randomly chosen capability will match the
password.
WEBIST 2007 - International Conference on Web Information Systems and Technologies
408
0
0.2
0.4
0.6
0.8
1
1.2
1234
Running time (z)
Input Values per time (T)
x=1
x=2
x=3
x=4
x=5
x=6
Figure 2: Inputs per time.
0
0.2
0.4
0.6
0.8
1
1.2
X=2 X=3 X=4 X=5 X=6
Passwords length
Inpute values per time (T)
Z=1
Z=2
Z=3
Z=4
Figure 3: Passwords length per input values.
The value of X also affect the validity time of
capability. The minimum validity time
()
min
m
is
()
min
m
=
()
1x .T
k
, where T
k
denotes the time a
key is typed in. The maximum validity
time
()
max =
.T
m
k
. Ideally, we would like to get
a small intervals for the validity time, so that we can
tightly control the validity period, so we would like
large value of X to minimize the difference between
()
min
m
and
()
max
m
. We can determine X from
Min
m
and Max
m
max
X=
max - min
m
mm
(4)
The
max
m
metric defines the longest amount
of time that passwords matches can remain idle and
still have a valid capability. Put in another way,
Max
m
defines the maximum amount of time that
an attacker can login with particular capability
before the capability is rejected by authentications.
Because the (login denial) rejection probabilities
are independent Bernoulli trials the probability that
the end user/client and server will be able to
establish correct password after one try (by
exchanging password is:
i
P (connect after 1 try) = (1 - )
i
The probability that client will connect after K tries
is:-
P (connection after K tries)
=
k
1 - (1 - P (connect after 1 try))
i k
= 1 - (1 - ) )
i
For a given desire connection probability
, P
(connect)
the required numbers of connection
attempts is:-
(
)
(
)
()
lo g 1 -
K=
i
lo g 1 - 1-
i
P connect
⎛⎞
⎜⎟
⎝⎠
(5)
A nice feature of this formula is that the expected
number of connections attempts depends
logarithmically on the matches’ probability, which
indicates that even for large
i
; a determined client
can get a connected after a moderate waiting time.
4 CONCLUSION
We have presented an algorithm that explains the
ticket message exchange for access to an end user in
distributed systems. Using this model as a basis, we
also present a countermeasure that may be deployed
in the current passwords used, assuming that client
and server software is updated.
Our approaches exploit the input capabilities of
login that allow us to determine inputs time from the
passwords length in which it occurs. We prove our
arguments to assess the security mathematically.
REFERENCES
Palmgren, K.,Diffie-Hellman Key Exchange – A Non-
Mathematician’s Explanation” February 2005.
Steiner et al., J.G., Kerberos: An Authentication Service
for Open Network Systems March 1988.
Ian Jermyn et al, “The Design and Analysis of Graphical
Passwords,” Proceeding of the 8
th
USENIX Security
Symposium August 1999].
Morris, R., and Thomson, K., “Password Security, A case
History,” Communications of the ACM 22 (11),
pp.594597 (Nov.1979).
Lamport, L., “Password Authentication with insecure
Communication, “Communication of the ACM 24
(11), pp.770771 (Nov. 1981).
END USER AUTHENTICATION (EUA)
409