Improving Resilience of Behaviometric Based Continuous Authentication with Multiple Accelerometers

hamme, Tim Van; Preuveneers, Davy; Joosen, Wouter

doi:10.1007/978-3-319-61176-1_26

Cited by 9 publications

(5 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sensor based gait recognition is regarded as a promising approach towards unobtrusive user authentication [9,17,29,30]. Despite being less robust than well-established biometrics, motion data takes advantage of body worn sensors that are widely implemented in modern devices and require little to no effort by the user.…”

Section: Introductionmentioning

confidence: 99%

Data Privatizer for Biometric Applications and Online Identity Management

Garofalo

Preuveneers

Joosen

2020

Privacy and Identity Management. Data for Better Living: AI and Privacy

Self Cite

View full text Add to dashboard Cite

Biometric data embeds information about the user which enables transparent and frictionless authentication. Despite being a more reliable alternative to traditional knowledge-based mechanisms, sharing the biometric template with third-parties raises privacy concerns for the user. Recent research has shown how biometric traces can be used to infer sensitive attributes like medical conditions or soft biometrics, e.g. age and gender. In this work, we investigate a novel methodology for private feature extraction in online biometric authentication. We aim to suppress soft biometrics, i.e. age and gender, while boosting the identification potential of the input trace. To this extent, we devise a min-max loss function which combines a siamese network for authentication and a predictor for private attribute inference. The multi-objective loss function harnesses the output of the predictor through adversarial optimization and gradient flipping to maximize the final gain. We empirically evaluate our model on gait data extracted from accelerometer and gyroscope sensors: our experiments show a drop from 73% to 52% accuracy for gender classification while loosing around 6% in the identity verification task. Our work demonstrates that a better trade-off between privacy and utility in biometric authentication is not only desirable but feasible.

show abstract

Section: Introductionmentioning

confidence: 99%

Data Privatizer for Biometric Applications and Online Identity Management

Garofalo

Preuveneers

Joosen

2020

Privacy and Identity Management. Data for Better Living: AI and Privacy

Self Cite

View full text Add to dashboard Cite

show abstract

“…However, sensor data from two entirely different body areas, e.g., foot versus hip, are fundamentally different. Small sensor displacements have shown to have an impact [34]. It is important to have training data sampled from the same underlying distribution as the data at test time.…”

Section: Gait Dataset and Featuresmentioning

confidence: 99%

“…Beyond the typical step counting and fall detection applications, gait information can offer good indicators to recognize certain health conditions, such as Parkinson’s disease [12,13]. In the area of security, the 3D acceleration and angular velocity in someone’s gait can be used as a soft biometric to authenticate someone by the way the walk [14,15,16]. Indeed, accelerometers and gyroscopes can offer a wealth of information to make applications smart.…”

Section: Introductionmentioning

confidence: 99%

A Systematic Comparison of Age and Gender Prediction on IMU Sensor-Based Gait Traces

hamme

Garofalo

Rúa

et al. 2019

Sensors

Self Cite

View full text Add to dashboard Cite

Sensors provide the foundation of many smart applications and cyber–physical systems by measuring and processing information upon which applications can make intelligent decisions or inform their users. Inertial measurement unit (IMU) sensors—and accelerometers and gyroscopes in particular—are readily available on contemporary smartphones and wearable devices. They have been widely adopted in the area of activity recognition, with fall detection and step counting applications being prominent examples in this field. However, these sensors may also incidentally reveal sensitive information in a way that is not easily envisioned upfront by developers. Far worse, the leakage of sensitive information to third parties, such as recommender systems or targeted advertising applications, may cause privacy concerns for unsuspecting end-users. In this paper, we explore the elicitation of age and gender information from gait traces obtained from IMU sensors, and systematically compare different feature engineering and machine learning algorithms, including both traditional and deep learning methods. We describe in detail the prediction methods that our team used in the OU-ISIR Wearable Sensor-based Gait Challenge: Age and Gender (GAG 2019) at the 12th IAPR International Conference on Biometrics. In these two competitions, our team obtained the best solutions amongst all international participants, and this for both the age and gender predictions. Our research shows that it is feasible to predict age and gender with a reasonable accuracy on gait traces of just a few seconds. Furthermore, it illustrates the need to put in place adequate measures in order to mitigate unintended information leakage by abusing sensors as an unanticipated side channel for sensitive information or private traits.

show abstract

“…smartwatch vs. smartphone). Even if sampling on multiple positions on the body can improve the equal error rate of gait analysis [19], the obtained behaviometrics will not be statistically independent. Last but not least, from a security point of view, one must consider the fact that certain behaviometrics can be captured in a public setting by malicious adversaries that can later use them to spoof the identify of a subject.…”

Section: Limitations Of Decision Fusion Schemesmentioning

confidence: 99%

Managing distributed trust relationships for multi-modal authentication

hamme

Preuveneers

Joosen

2018

Journal of Information Security and Applications

Self Cite

View full text Add to dashboard Cite

Multi-modal active authentication schemes fuse decisions of multiple behavioral biometrics (behaviometrics) to reduce identity verification errors. The challenge that we address in this work is the security risk caused by these decision fusion schemes making invalid assumptions, such as a fixed probability of (in)correct recognition and a temporal congruence of behaviometrics. To mitigate this risk, this paper presents a formal trust model that drives the behaviometric selection and composition. Our trust model adopts a hybrid approach combining policy and reputation based knowledge representation techniques. Our model and framework (1) externalizes trust knowledge from the authentication logic to achieve loosely coupled trust management, and (2) formalizes this knowledge in description logic to reason upon and broker complex distributed trust relationships to make risk-adaptive decisions for multi-modal authentication. The evaluation of our proof-of-concept illustrates an acceptable performance overhead while lifting the burden of manual trust and behaviometric management for multi-modal authentication.

show abstract

Improving Resilience of Behaviometric Based Continuous Authentication with Multiple Accelerometers

Cited by 9 publications

References 23 publications

Data Privatizer for Biometric Applications and Online Identity Management

Data Privatizer for Biometric Applications and Online Identity Management

A Systematic Comparison of Age and Gender Prediction on IMU Sensor-Based Gait Traces

Managing distributed trust relationships for multi-modal authentication

Contact Info

Product

Resources

About