Privacy-Protecting Techniques for Behavioral Biometric Data: A Survey

Abstract

Our behavior -the way we talk, walk, or think-is unique and can be used as a biometric trait. It also correlates with sensitive attributes like emotions. Hence, techniques to protect individuals' privacy against unwanted inferences are required. To consolidate knowledge in this area, we systematically reviewed applicable anonymization techniques. We taxonomize and compare existing solutions regarding privacy goals, conceptual operation, advantages, and limitations. Our analysis shows that some behavioral traits (e.g., voice) have received much attention, while others (e.g., eye-gaze, brainwaves) are mostly neglected. We also find that the evaluation methodology of behavioral anonymization techniques can be further improved.CCS Concepts: • Security and privacy → Pseudonymity, anonymity and untraceability.

Private user-centric management of electronic services in smart communities

“…The authors survey anonymization methods that they identified in the literature analysis to mitigate the two main identified privacy threats, namely identity disclosure to identify the user in another scenario and attribute disclosure to derive sensitive attributes form the behavioural biometrics. They give in the actualized paper version [73] of [24] an overview table indicating for the related privacy goals identity and attribute protection the different techniques that try to achieve these goals.…”

“…The authors grounded their work on "two main privacy threats that apply to behavioural data collected/processed by a third party", identity disclosure and attribute disclosure, which are in line with the PB3 Unlinkable and PB4 Resilient-to-Identifiability that we defined. In the actualized paper version [73], the authors provide an overview table indicating the privacy goals that the different techniques try to achieve.…”

“…In [73] the extended version 2 of [24] the authors sum up their survey with an overview in two tables. One table provides an overview of all found methods classified by trait and method, while a second table offers an overview over which privacy goals the different techniques try to achieve.…”

“…The identity protection comprises transformation of behavioural biometric data so that a person cannot be linked to the data. This includes pseudonymization and anonymization in relation to the identity [73]. The attribute protection comprises transformation of behavioural biometric data to protect specific private attributes, up to template protection, which is then still usable for authentication [24,73].…”

“…This includes pseudonymization and anonymization in relation to the identity [73]. The attribute protection comprises transformation of behavioural biometric data to protect specific private attributes, up to template protection, which is then still usable for authentication [24,73]. All traits can be used for identity and attribute inference, and thus to link user to data, identity theft or private attribute (e.g.…”