Continuous User Authentication Based on Keystroke Dynamics through Neural Network Committee Machines

This chapter proposes an authentication methodology that is both inexpensive and non-intrusive and authenticates users continuously while using a computer keyboard. This proposed methodology uses neural network committee machines. The committee consists of several independent neural networks trained to recognize a behavioral biometric characteristic: user’s typing pattern. Continuous authentication prevents potential attacks when users leave their desks without logging out or locking their computer session. Some experiments were conducted to evaluate and to calibrate the authentication committee. Best results show that a 0% FAR and a 0.15% FRR can be achieved when different thresholds are used in the system for each user. In this proposed methodology, capture system does not need to concern about typing errors in the text. Another feature of this methodology is that new users can be easily added to the system, with no need to re-train all neural networks involved.

Keystroke Analysis as a Tool for Intrusion Detection

In this chapter we discuss the potentialities of Keystroke Analysis as a tool for Intrusion Detection and other security applications, and investigate experimentally how the accuracy of the analysis scales with the increase of the number of individuals involved, a fundamental issue if we want to add Keystroke Analysis to the set of tools that can be used to improve the security of our computers and networks.

Cognitive Biometrics

Cognitive biometrics is a new authentication scheme that utilises the cognitive, emotional, and conative state of an individual as the basis of user authentication and/or identification. These states of mind (and their derivatives) are extracted by recording various biosignals such as the EEG, ECG, and electrodermal response (EDR) of the individual in response to the presentation of the authentication stimulus. Stimuli are selected which elicit characteristic changes within the acquired biosignal(s) that represent unique responses from the individual. These characteristic changes are processed using a variety of machine learning algorithms, resulting in a unique signature that identifies or authenticates the individual. This approach can be applied in both static mode (single point of authentication), or in continuous mode, either alone, or in a multi-modal approach. The data suggest that the classification accuracy can reach 100% in many scenarios, providing support for the efficacy of this new approach to both static and continuous biometrics.

Performance Metrics and Models for Continuous Authentication Systems

Continuous Authentication (CA) systems represent a new class of security systems that are increasingly the focus of much attention in the research literature. CA departs from the traditional (static) authentication scheme by repeating several times the authentication process dynamically throughout the entire login session; the main objectives are to detect session hijacking and ensure session security. As the technology gains in maturity and becomes more diverse, it is essential to develop common and meaningful evaluation metrics that can be used to compare and contrast between existing and future schemes. So far, all the CA systems proposed in the literature were by default evaluated using the same accuracy metrics used for static authentication systems. As an alternative, we discuss in this chapter dynamic accuracy metrics that better capture the continuous nature of CA activity. Furthermore, we introduce and study diverse and more complex forms of the Time-To-Authenticate (TTA) metrics corresponding to the authentication delay. We study and illustrate empirically the proposed metrics and models using a combination of real and synthetic data samples.

Low Level Multispectral Palmprint Image Fusion for Large Scale Biometrics Authentication

Continuous biometric authentication is a process where the installed biometric systems continuously monitor and authenticate the users. Biometric system could be an exciting application to log in to computers and in a network system. However, due to malfunctioning in high-security zones, it is necessary to prevent those loopholes that often occur in security zones. It has been seen that when a user is logged in to such systems by authenticating to the biometric system installed, he/she often takes short breaks. In the meantime some imposter may attack the network or access to the computer system until the real user is logged out. Therefore, it is necessary to monitor the log in process of the system or network by continuous authentication of users. To accomplish this work we propose in this chapter a continuous biometric authentication system using low level fusion of multispectral palm images where the fusion is performed using wavelet transformation and decomposition. Fusion of palmprint instances is performed by wavelet transform and decomposition. To capture the palm characteristics, a fused image is convolved with Gabor wavelet transform. The Gabor wavelet feature representation reflects very high dimensional space. To reduce the high dimensionality, ant colony optimization algorithm is applied to select relevant, distinctive, and reduced feature set from Gabor responses. Finally, the reduced set of features is trained with support vector machines and accomplishes user recognition tasks. For evaluation, CASIA multispectral palmprint database is used. The experimental results reveal that the system is found to be robust and encouraging while variations of classifiers are used. Also a comparative study of the proposed system with a well-known method is presented.

Towards Continuous Authentication Based on Gait Using Wearable Motion Recording Sensors

In this chapter we present continuous authentication using gait biometric. Gait is a person’s manner of walking and gait recognition refers to the identification and verification of an individual based on gait. This chapter discusses advantages and disadvantages of gait biometrics in the context of continuous authentication. Furthermore, we present a framework for continuous authentication using gait biometrics. The proposed framework extends on traditional static (one-time) user authentication. The framework can also be applied to other biometric modalities with small modifications.

Introduction to Continuous Authentication

Continuous Authentication (CA) systems represent a new generation of security mechanisms that continuously monitor user behavior and use this as basis to re-authenticate periodically throughout a login session. CA has been around for about a decade. As a result a limited amount of research work has been produced to date, and the first commercial products have only recently started reaching the market. We attempt, in this chapter, to provide some general perspectives in order to help achieve some common and better understanding of this emerging field. The chapter introduces basic CA concepts and terminologies, discusses the characteristics of CA data sources, and identifies major areas of application for CA systems.

Sitting Postures and Electrocardiograms

This approach is novel in terms of “participation” – the driver has neither to operate something nor to attach a device. Furthermore he/she must not be aware of the continuous collection of his/her personal profile at all.

Multimodal Biometric Hand-Off for Robust Unobtrusive Continuous Biometric Authentication

Conventional access control solutions rely on a single authentication to verify a user’s identity but do nothing to ensure the authenticated user is indeed the same person using the system afterwards. Without continuous monitoring, unauthorized individuals have an opportunity to “hijack” or “tailgate” the original user’s session. Continuous authentication attempts to remedy this security loophole. Biometrics is an attractive solution for continuous authentication as it is unobtrusive yet still highly accurate. This allows the authorized user to continue about his routine but quickly detects and blocks intruders. This chapter outlines the components of a multi-biometric based continuous authentication system. Our application employs a biometric hand-off strategy where in the first authentication step a strong biometric robustly identifies the user and then hands control to a less computationally intensive face recognition and tracking system that continuously monitors the presence of the user. Using multiple biometrics allows the system to benefit from the strengths of each modality. Since face verification accuracy degrades as more time elapses between the training stage and operation time, our proposed hand-off strategy permits continuous robust face verification with relatively simple and computationally efficient classifiers. We provide a detailed evaluation of verification performance using different pattern classification algorithms and show that the final multi-modal biometric hand-off scheme yields high verification performance.

Personal Identification and Authentication Based on Keystroke Dynamics in Japanese Long-Text Input

We have investigated several characteristics of keystroke dynamics in Japanese long-text input. We performed experiments with 189 participants, classified into three groups according to the number of letters they could type in five minutes. In this experimental study, we extracted feature indices from the keystroke timing for each alphabet letter and for each two-letter combination composed of a consonant and vowel in Japanese text. Taking into account two identification methods using Weighted Euclidean Distance (WED) and Array Disorder (AD), we proposed a hybrid model for identifying individuals on the basis of keystroke data in Japanese long-text input. By evaluating the identification performance of individuals in the three groups, the effectiveness of the method was found to correspond to the typing skill level of the group.