Information Technology Reference
In-Depth Information
Algorithm. Enrollment
V ← ø;
for i = t+1 to r do
x i , y i U F - T;
V
← T (x i , y i );
Output V;
After enrollment, the server stores vaults in a database. At this time, since the
minutiae points in vault are transformed by H(p), if an attacker has vault, he just gets
transformed minutiae. Therefore, it is secure against ARM and SKI attack. In authen-
tication process, the server decodes the vault, and then compares transmitted minutiae
with stored minutiae. If both of minutiae are same, user can be authenticated. We
denote (x' i , y' i ) (b i , 0) J overlapping of J onto the x-coordinate bi. If the result of
CRCcode is positive, the algorithm outputs “match”. It the result of CRCcode is nega-
tive, the algorithm outputs “non-match”.
Public parameters: A field F , a Cyclic Redundancy
Check code CRCcode.
Input: Parameter t, r such that t r. A secret S F k ,
A set B={b i } t i=1 , where b i F
Output: A secret S' F k
{'null'}
Algorithm. Authentication
Q ← ø;
J ← Q
V
for i = 1 to t do
(x' i , y' i ) (bi, 0) J;
Q ← Q (x' i , y' i );
D ← CRCcode(Q);
if D is positive
output “match”
else
output “non-match”
To protect the user's fingerprint, we also proposed one thing. K.Nandakumar de-
scribes the equation (1) that transformed minutiae using plus operation between the
scan image and password. We think that should just one operation. It is not only an
addition but a minus, a multiplication, a division in the equation. If the operations
apply each quadrant using arbitrary rules or criterions, the user's biometric informa-
tion and password will be protected more than from attack of malicious person.First, a
fingerprint of user divided into quadrant. The minutiae that extracted from the user's
fingerprint transformed according to the operation by the arbitrary rules or criterions
in each quadrant. Or, the minutiae could be transformed according to the operation by
the arbitrary rules or criterions when fingerprint of arbitrary user be enrolled.
For example, we assume a fingerprint of user divided into quadrant. The minutiae
that extracted from the user's fingerprint transformed according to the operation by
the arbitrary rules or criterions in each quadrant. The scanned fingerprint of user will
be extracted and correspond to the quadrant. In this time, we describe a point of refer-
ence that according to the arbitrary rules or criterions. The upper and lower parts of
this point are applied difference operations and then the minutiae will be transformed.
Search WWH ::




Custom Search