Using recurrent neural networks for continuous authentication through gait analysis

Abstract This letter presents a novel framework for continuous user authentication of mobile devices based on gait analysis, exploiting inertial sensors and Recurrent Neural Network for deep-learning based classification. The proposed framework handles all the stages of the continuous authentication, starting from data collection, to data preprocessing, classification and policy enforcement. The letter will puts its emphasis on the data analysis aspects, discussing the methodologies used to improve the quality of classification, including data augmentation and a sliding window interval approach for improved training. Furthermore, it will be discussed the enforcement which is based on the Usage Control paradigm for continuous policy enforcement. A set of real experiments will demonstrate the effectiveness and efficiency of the proposed framework.

[1]  Judith Liu-Jimenez,et al.  Gait recognition using smartphone , 2016, 2016 IEEE International Carnahan Conference on Security Technology (ICCST).

[2]  Dana Kulic,et al.  Data augmentation of wearable sensor data for parkinson’s disease monitoring using convolutional neural networks , 2017, ICMI.

[3]  Min Zhang,et al.  Gait Activity Authentication Using LSTM Neural Networks with Smartphone Sensors , 2019, 2019 15th International Conference on Mobile Ad-Hoc and Sensor Networks (MSN).

[4]  N. Arunkumar,et al.  Gait and tremor investigation using machine learning techniques for the diagnosis of Parkinson disease , 2018, Future Gener. Comput. Syst..

[5]  Faicel Chamroukhi,et al.  Physical Human Activity Recognition Using Wearable Sensors , 2015, Sensors.

[6]  Davide Anguita,et al.  A Public Domain Dataset for Human Activity Recognition using Smartphones , 2013, ESANN.

[7]  Jaehong Park,et al.  The UCONABC usage control model , 2004, TSEC.

[8]  Romain Tavenard,et al.  Data Augmentation for Time Series Classification using Convolutional Neural Networks , 2016 .

[9]  Thaier Hayajneh,et al.  Smartphone and Smartwatch-Based Biometrics Using Activities of Daily Living , 2019, IEEE Access.

[10]  Waleed Ejaz,et al.  ADLAuth: Passive Authentication Based on Activity of Daily Living Using Heterogeneous Sensing in Smart Cities , 2019, Sensors.

[11]  Qing Yang,et al.  HMOG: New Behavioral Biometric Features for Continuous Authentication of Smartphone Users , 2015, IEEE Transactions on Information Forensics and Security.

[12]  David Mohaisen,et al.  AUToSen: Deep-Learning-Based Implicit Continuous Authentication Using Smartphone Sensors , 2020, IEEE Internet of Things Journal.

[13]  Qian Wang,et al.  Deep Learning-Based Gait Recognition Using Smartphones in the Wild , 2018, IEEE Transactions on Information Forensics and Security.

[14]  Hesham H. Ali,et al.  Machine Learning and Similarity Network Approaches to Support Automatic Classification of Parkinson's Diseases Using Accelerometer-based Gait Analysis , 2019, HICSS.

[15]  Gianluca Dini,et al.  Risk analysis of Android applications: A user-centric solution , 2018, Future Gener. Comput. Syst..

[16]  Fabio Martinelli,et al.  Walking Through the Deep: Gait Analysis for User Authentication Through Deep Learning , 2018, SEC.