Abstract:Smartphones are equipped with a set of sensors that describe the environment (e.g., GPS, noise, etc.) and their current status and usage (e.g., battery consumption, accelerometer readings, etc.). Several works have already addressed how to leverage such data for user-in-a-context continuous authentication, i.e., determining if the porting user is the authorized one and resides in his regular physical environment. This can be useful for an early reaction against robbery or impersonation. However, most previous … Show more
“…In addition, the energy consumption of their system is higher than the consumption of ours. On the other hand, the work proposed by de Fuentes et al [ 21 ] performs a user classification based on non-assisted sensors achieving 97% of accuracy using only battery reading information. When the system tries to identify both user and environment by combining data from different sensors, like battery readings, ambient light and ambient noise sensors, it obtains 81.35% of accuracy.…”
Section: Discussionmentioning
confidence: 99%
“…The ML method selected was One-Class SVM [ 19 ] after comparing it to others like Kernel Ridge Regression (KRR) [ 20 ] and k-Nearest Neighbours (kNN). Another interesting solution is the proposed by de Fuentes et al [ 21 ]. The authors of this work used non-assisted sensors, such as battery, transmitted data, ambient light and noise to authenticate the user.…”
Continuous authentication systems for mobile devices focus on identifying users according to their behaviour patterns when they interact with mobile devices. Among the benefits provided by these systems, we highlight the enhancement of the system security, having permanently authenticated the users; and the improvement of the users’ quality of experience, minimising the use of authentication credentials. Despite the benefits of these systems, they also have open challenges such as the authentication accuracy and the adaptability to new users’ behaviours. Continuous authentication systems should manage these challenges without forgetting critical aspects of mobile devices such as battery consumption, computational limitations and response time. With the goal of improving these previous challenges, the main contribution of this paper is the design and implementation of an intelligent and adaptive continuous authentication system for mobile devices. The proposed system enables the real-time users’ authentication by considering statistical information from applications, sensors and Machine Learning techniques based on anomaly detection. Several experiments demonstrated the accuracy, adaptability, and resources consumption of our solution. Finally, its utility is validated through the design and implementation of an online bank application as proof of concept, which allows users to perform different actions according to their authentication level.
“…In addition, the energy consumption of their system is higher than the consumption of ours. On the other hand, the work proposed by de Fuentes et al [ 21 ] performs a user classification based on non-assisted sensors achieving 97% of accuracy using only battery reading information. When the system tries to identify both user and environment by combining data from different sensors, like battery readings, ambient light and ambient noise sensors, it obtains 81.35% of accuracy.…”
Section: Discussionmentioning
confidence: 99%
“…The ML method selected was One-Class SVM [ 19 ] after comparing it to others like Kernel Ridge Regression (KRR) [ 20 ] and k-Nearest Neighbours (kNN). Another interesting solution is the proposed by de Fuentes et al [ 21 ]. The authors of this work used non-assisted sensors, such as battery, transmitted data, ambient light and noise to authenticate the user.…”
Continuous authentication systems for mobile devices focus on identifying users according to their behaviour patterns when they interact with mobile devices. Among the benefits provided by these systems, we highlight the enhancement of the system security, having permanently authenticated the users; and the improvement of the users’ quality of experience, minimising the use of authentication credentials. Despite the benefits of these systems, they also have open challenges such as the authentication accuracy and the adaptability to new users’ behaviours. Continuous authentication systems should manage these challenges without forgetting critical aspects of mobile devices such as battery consumption, computational limitations and response time. With the goal of improving these previous challenges, the main contribution of this paper is the design and implementation of an intelligent and adaptive continuous authentication system for mobile devices. The proposed system enables the real-time users’ authentication by considering statistical information from applications, sensors and Machine Learning techniques based on anomaly detection. Several experiments demonstrated the accuracy, adaptability, and resources consumption of our solution. Finally, its utility is validated through the design and implementation of an online bank application as proof of concept, which allows users to perform different actions according to their authentication level.
“…The work presented in [ 11 ] studied the utility of information representing battery consumption, transmitted data, and background noise and light (and combinations of them) for CA. The information, collected from the SherLock database, permits the device to work autonomously on the CA process, as it is non-assisted sensorial data.…”
Ensuring the confidentiality of private data stored in our technological devices is a fundamental aspect for protecting our personal and professional information. Authentication procedures are among the main methods used to achieve this protection and, typically, are implemented only when accessing the device. Nevertheless, in many occasions it is necessary to carry out user authentication in a continuous manner to guarantee an allowed use of the device while protecting authentication data. In this work, we first review the state of the art of Continuous Authentication (CA), User Profiling (UP), and related biometric databases. Secondly, we summarize the privacy-preserving methods employed to protect the security of sensor-based data used to conduct user authentication, and some practical examples of their utilization. The analysis of the literature of these topics reveals the importance of sensor-based data to protect personal and professional information, as well as the need for exploring a combination of more biometric features with privacy-preserving approaches.
“…The heartbeat signal can be used as a unique feature to authenticate smartphone users. In [25], the author explored four mobile phone non assisted sensors; transmitted data, noise, battery and ambient light to develop a continuous user authentication based on KNN. The KNN classifier achieved a reasonable accuracy.…”
The ever-growing technology in mobile smartphones has enabled users to store sensitive and private information; as a result, it required the need for an improved security system. Previous approaches heavily relied on shallow machine learning algorithms that require feature extraction which is time consuming, laborious and can cause, resulting in poor authentication. In this paper, we propose a deep learning-dense neural network to avoid the limitation of the classical algorithms and build a mobile smartphone touch screen authentication scheme based on keystroke dynamics. A deep learningdense neural network classifier was trained using keystroke dynamics features extracted from users. A comparative analysis was made between our proposed DNN classifier and some selected classical machine learning algorithms on the keystroke dynamics data. The data is split into five different data partition of training and testing. Results clearly indicated that the deep learningdense neural network has eliminated the feature extraction steps required by the classical algorithms and improved the overall authentication accuracy, as such, improved the security of the smartphone device. In addition, it is found that the propose deep learningdense neural network authentication scheme is more robust than the classical algorithms and has the potential to be fully implemented on smartphone to improve the security system of the mobile smartphone touch screen devices.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.