Communications: electrical – Selective – Intelligence comparison for controlling
Reexamination Certificate
2000-10-13
2004-11-16
Horabik, Michael (Department: 2635)
Communications: electrical
Selective
Intelligence comparison for controlling
C340S005820, C713S186000
Reexamination Certificate
active
06819219
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of Invention
The present invention relates generally to the field of biometric-based authentication. More specifically, the present invention is related to biometric-based authentication in wireless communications for access control.
2. Discussion of Prior Art
In an age where electronic transactions are prevalent, safe-guarding data has become an important issue, thus giving rise to a myriad of security systems. Two of the common security systems used are password and personal identification (PIN) systems. Password systems require a user to provide the authentication system with a username and a password (both of which are unique to the user.) PIN systems on the other hand usually require a user to provide a code, usually referred to as the PIN code, for authentication purposes. Both the password and the PIN system can prove to be a nuisance to users in the event they forget their password or PIN code. Moreover, a user A can easily impersonate another user B if user A happens to get a hold of the password or PIN code (given either voluntarily or cracked through other means) of user B. One way to avoid such breaches of security is to implement a user-based physiological or behavioral characteristics as a means for authentication. This is the general idea behind biometrics.
Biometrics is the study of measurable biological characteristics. In computer security, biometrics refers to authentication techniques that rely on measurable physiological or behavioral characteristics that can be automatically checked.
In the above description, authentication is usually accomplished via a biometric device. A general description of the functionality of a biometric device now follows. First, the biometric device captures a profile of the characteristic and next, a comparison of the acquired profile is made with a stored profile or template. Lastly, upon successful matching of the captured and stored profile, the user is interfaced with the application system requesting authentication.
Biometrics, as described above and in
FIG. 1
, depend on physiological
104
or behavioral
106
characteristics of a person. Physiological characteristic
104
is a stable physical characteristic associated with a person. In other words, it is a set of physical characteristics (of a person) that does not change, in any considerable way, with respect to time. Some examples of physiological characteristics
104
that are stable and which can be utilized for authentication purposes include, but are not limited to: fingerprint pattern
108
, retinal pattern
110
, iris pattern
112
, facial pattern
114
, hand silhouette pattern
116
, or blood vessel pattern on the back of the eye.
On the other hand, a behavioral characteristic
106
involves a person's psychological makeup in conjunction with various other general physical traits such as sex and size. Examples of behavioral characteristics
106
that can be monitored include, but are not limited to: signature pattern
118
, voice pattern
120
, or typing pattern or keystroke dynamics
122
. A general description of some of the physiological and behavioral characteristics are outlined below.
Authentication based on fingerprint: One of the most common biometric techniques, wherein users scan in a copy of their fingerprint and a comparison is performed by the authentication device as to whether or not the input fingerprint matches that of a stored fingerprint, corresponding to the same person. Some fingerprint authentication devices take it a step further and check for a pulse to combat problems posed by false-authentication via fingerprints that are not real.
Authentication based on hand geometry: An authentication querying system captures the physical characteristics of a user's hand and fingers via a scanner and is matched with a stored template of the same user. Upon successful authentication, an action (like opening a secure door) is performed by the querying system.
Authentication based on retinal scanning: A scanner scans at close range a user's retina (the image forming innermost coat of the black part of the eye ball) using a low intensity light, creates a eye signature, matches it with a stored retinal template, and performs a specific action upon successful authentication. It should however be noted that failure of a user to focus correctly results in inaccurate results.
Authentication based on iris scanning: An iris scanner scans unique random patterns of the iris (the colored part of the eye) and authenticates users based on comparing the consistency of the acquired pattern with that of stored patterns. Unlike retinal scanning, close range interaction is not required.
Authentication based on facial recognition: A facial recognition system scans (the features of a users face) and captures an image of the user's face and compares it to a stored static facial image of the same user. Upon successful authentication, a specific action is performed by the facial recognition system.
Authentication based on signature verification: This authentication technique utilizes a pressure sensitive pen and a tablet to record a user's signature. The system then compares it against stored samples of signatures corresponding to the same user, and upon authentication, performs a specific action.
Authentication based on voice recognition: Authentication in this technique is based on recognizing voice and speech characteristics (associated with a user) that are imperceptible and hence not replicable. Voice recognition systems typically require more memory for storing voice templates of users.
Hence, biometrics are beginning to play a critical role in authentication and security. Biometrics authenticate the user not based on what he can remember (like passwords, PIN's, etc.), but rather use the user's characteristics (or who the user is) to perform authentication.
FIG. 2
illustrates prior art biometric authentication system
200
. A combination of two authentication scenarios are illustrated in this example. In the first scenario user
201
uses smart card
202
, and in the second scenario, a scan for an biometric attribute is performed
208
, on user
201
, for further authentication. In the first scenario, user
201
inserts smart card
202
into smart card reader system
204
. Then, the smart card reader system
204
extracts biometric profile
206
(stored in smart card
202
) that is unique for user
201
. Next, the extracted profile
206
is compared with profiles stored in database
212
to determine if a match exists. Furthermore, in the second scenario, a scan for attribute
208
(e.g., a retinal scan) is performed on user
201
and the system creates digital profile
210
of the measured attribute. Next, the measured digital profile is compared with digital profiles stored in database
212
to determine if a match exists, and upon successful matching, an action is performed (like opening a secure door).
The prior art scenario described above exposes some of the pitfalls associated with these authentication systems. A biometric template or profile associated with a user, as described in the above example, is either stored on smart card (first scenario)
202
, which must be inserted into reader or scanner
204
, or on server (second scenario)
212
. One common problem associated with the first scenario is that smart card reader
204
in the authentication system processes information in a very slow and time consuming manner. Each time a card is removed from the user and placed in a reader, increases the chance that the card will be left in the reader or otherwise lost. The exposed contacts of smart cards and readers make them susceptible to dirt, grime and other contaminants, decreasing reliability. This is particularly an issue for outdoor verification, for example at automatic teller machines which are exposed to weather and vandalism.
One common concern associated with the second scenario is that storing the biometric profile in database
212
and accessing it over a network is not completely
Bolle Rudolf Maarten
Nunes Sharon Louise
Pankanti Sharathchandra
Ratha Nalini Kanta
Smith Barton Allen
Horabik Michael
Lacasse Randy W.
Lacasse & Associates LLC
Nguyen Nam
Soundararajan Ramraj
LandOfFree
Method for biometric-based authentication in wireless... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for biometric-based authentication in wireless..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for biometric-based authentication in wireless... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3275990