Authors
Sulavko Alexey E.
|
Degree
|
Cand. Sci. (Eng.), Assistant Professor, Integrated Information Security Department, Omsk State Technical University |
|---|---|
|
E-mail
|
sulavich@mail.ru |
|
Location
|
Omsk, Russia |
|
Articles
|
Two-factor authentication of users of computer systems on remote server using the keyboard handwritingauthentication
server. The method of two-factor authentication of users of computer systems on the remote
server using personal biometric data is proposed. The method based on error-correcting coding and
other conversion of biometric data. The developed method is based on «fuzzy extractors» and allows
to store only fragments of biometric standard on the server and does not allow to restore the standard
if this fragments were stolen. As the biometric features of a person is proposed to use the keystroke
dynamics: duration of retention and the time intervals between keystrokes as a person type the passphrase
on the keypad. An original way to use information about the stability of biometric features is
proposed. The information about biometric features stability is used to choose the best ones for preparing
a cryptographic key and decrease errors of key generation. Also it is a part of a secret information
that storages on the server side and used in key recovery procedure. As a part of the future research
for «combining» and «subtraction» bit sequences of PRN code and biometric data for cryptographic
key generation it is planned to use fuzzy implication operation, adapting one of the fuzzy inference
algorithms (Tsukamoto, Sugeno, Mamdani, Larsen et al.)
Read more...
Generation of key sequences based on voice messagesThe problem of the generation of the key sequences on the basis of biometric data is described. Objective: To develop a method of generating a key sequence based on the subject of voice parameters with indicators of reliability and key length exceeding achieved. Two features spaces of human voice are proposed: dependent and independent of the uttered phrase. The methods of generating keys based on voice messages on the basis of fuzzy extractors using Hadamard or Bose — Chaudhuri — Hocquenghem error correcting codes are proposed. Also the ranking procedure of most stable features individual for each subject was proposed. The effectiveness of the proposed method was defined. The optimum methods for each proposed feature space have been found. These results are superior to previously achieved by generating a key sequence based on voice. Read more... Identification potential of keyboard handwriting considering vibration parameters and force keystrokesThe article considers the problem of data protection from unauthorized access by means of user identification
by keyboard handwriting. The estimation of informativeness of different features that characterize
the keyboard handwriting of subjects, including the dynamics of change in pressure when you
press the keys and keyboard settings vibration. The category of new features, based on using of wavelet
transform Daubechies D6 to function of the pressure fingers on the keys and keyboard functions of
vibration while typing, was proposed. The laws of distribution of basic and additional features of keyboard
handwriting were determined. To form the base of biometric samples a keyboard was designed
with the use of special sensors. The estimation of the correlation dependence of features was made. It is
determined that the correlation between basic features (temporal characteristics of keystrokes) and additional
features (pressure on the keys and the keyboard vibration) in more than 80% of cases is weak.
Thus, in the proposed new attributes contain information about the subject. An assessment of the probability
of identification errors based on the Bayesian strategy using the various features of the spaces was
made. It is found that additional features can reduce the average number of errors is more than 7 times.
Read more...
Testing of neurons based on statistical functionals for verifying biometric images in feature space with different informativeness
The scale of features informativeness is formulated and presented. The efficiency of «wide» neural
network neurons based on the various variations of following criteria for checking the distribution law
of a random variable was evaluated: Smirnov-Kramer-von Mises, Anderson-Darling, Watson, Frotsini,
average geometric comparison functions for probability densities Kolmogorov-Smirnov, Cooper. A criterion
for the maximum area of intersection of the comparable probability density functions is proposed.
Variants for the modernization of functionals for the processing of features with a noticeable and high
correlation dependence are found, in particular, on the basis of the Smirnov-Kramer-von Mises criterion.
The analysis of the results of the work made it possible to determine the most suitable conditions
for the use of the functionals under consideration, depending on the features of the feature space. The
boundary conditions of the experiment correspond to the features of dynamic biometric patterns. First
of all, the research was aimed at solving the problems of recognizing secret biometric images, but the
results obtained can be used in any verification task, where there are dependent and independent signs
of average informativeness, as well as little informative and extremely little informative features with
a distribution of values close to normal.
Read more...
Biometric authentication of information systems users using keyboard handwriting based on immune network algorithmsThe article deals with the problem of protecting information from the threat of unauthorized access
(theft, destruction, modification). A method of biometric authentication by keystroke dynamics using
an artificial immune system is proposed. Keystroke dynamics is a characteristic dependent on the
time and state of the operator. A feature of the approach is the use of a new architecture of computing
elements (detectors) at the core of the immune system. The detector is a separate measure of proximity,
which determines the distance to the template in the space of only a part of the possible features.
All detectors differ in “interaction interfaces” with a recognizable sample (a set of processed features)
and the principle of sample processing (each detector can be based on a separate measure of proximity).
In the aggregate, the detectors verify the biometric samples for belonging to the classes “Own”
or “Alien”. The developed artificial immune system (network) is capable of self – learning, each authentication
increases the reliability of its solutions. Also, an artificial immune network can determine
the degree of aging of the template and the fact that the user is authenticated in an altered functional
state. This information can be used in the implementation of access control procedures. The experimentally
achieved reliability indicators for solutions EER = 1.5–4% (depending on the state of the
subject). The comparison of the obtained result with the world level in this area has been carried out.
Read more...
Biometric authentication method based on cepstral characteristics of external ear echograms and biometrics-to-code neural converterOpen biometric images (fingerprint, iris, face) are "in sight" and therefore compromised in the natural environment. In this work, it is proposed to use data on the internal structure of the outer ear obtained using echography as biometric images. The individual characteristics of the ear canal of subjects are hidden from direct observation and cannot be copied by photographing. The proposed authentication method is based on cepstral analysis of echograms of the ear canal using neural network biometrics to code converters, trained in accordance with GOST R 52633.5. The neural network biometrics-code converter allows you to associate a user's cryptographic key or password with his biometric image. This is a shallow neural network of one or two layers of neurons, which is configured to generate a key specified during training when an image of a known user arrives, and when an unknown image arrives at its inputs, generate a random code with high entropy. At the entrance to this network, cepstral signs of echograms were received. To apply the method in practice, you need a special device that combines a headphone with a sound-proof housing and a microphone. The results obtained can be called optimistic EER = 0.031 (FAR = 0.001 at FRR = 0.23). The use of neural network converters biometrics-code showed a relatively higher percentage of errors in comparison with multilayer neural networks and the naive Bayes classification scheme, however, neural network biometrics to code converters allows you to implement authentication in a protected mode. This means that the subject's biometric data will be protected from compromise at the stages of storage, execution and transmission via communication channels. Read more... Verification of the personality of subjects by face based on neural network algorithms of artificial intelligence performed in protected modeThe aim of the work is to develop a system for verifying subjects by face based on a neural network model that is executed in a protected mode. Protected mode means that the identity verification system is highly resistant to destructive influences, such as competitive attacks, and allows storing and processing biometric data without compromising it. The system is based on a biometrics to code converter trained according to GOST R 52633.5, which allows you to associate the subject’s facial biometrics image with its cryptographic key or long password, which can later be used for authentication, and deep convolutional neural networks. For face detection in the image, the MTCNN artificial neural network architecture was used, and several neural network architectures were tested for feature extraction: InceptionResnet, Facenet512, VGG-Face and OpenFace. The best results were shown by the InceptionResnet neural network. When evaluating the effectiveness and testing the reliability of the proposed system on a special dataset of faces collected under different lighting conditions in a room, it was possible to achieve a relatively low value of equal probability of errors of the first and second kind (EER = 0.0146 with a key length of 278 bits), which confirms the effectiveness of the considered approach to building face verification systems. Read more... |
