Ubiquitous perception – by sensing of objects, subjects and gestures – is a pivotal opportunity for future technology: it enables personalized services such as smart living, automated logistics or interaction through free-space gestures. However, it also challenges ethical and moral boundaries and threatens privacy. Particularly, the charter of
fundamental rights of the EU (CFR-EU) highlights that balanced and sustainable development requires the protection of fundamental rights (art. 8: everyone should have the right to protect personal data). Implementation of article 8 into technology will enforce personal data protection as a fundamental right (GDPR).
HOLDEN proposes a radically new approach to perception by concisely analysing ethical constraints and privacy risks while re-thinking Radio Frequency (RF)-based sensing. The technology has a transformative positive effect on our economy and society by implementing article 8 of the CFR-EU through establishing necessary conditions for privacy and ethically compliant perception and derives adaptive sensing tools able to respect these constraints.
HOLDEN constitutes the first concentrated effort to explore social aspects of RF-sensing to guide technological advance towards compliance of ethical and privacy constraints (WP2): we develop a model of appropriation, to obtain better understanding of public acceptance of novel sensing-based technology by stimulating the public
awareness, considering social interactions, privacy threats and ethical implications.
From these findings we derive privacy and ethically compliant concepts for RF-based perception. In a radical paradigm shift from a technology-centric perspective to a privacy-centric perspective, we develop a distributed multiantenna system for simultaneous multi-target recognition, which is private by design. HOLDEN achieves this goal along three high-risk, complementary, and privacy-centric paths:
- Path 1: Continuous-space measurement points (WP3):
Ethically compliant radio-based 3D vision by holographic image processing of RF wave-fronts.
- Path 2: Discrete-space measurement points (WP4):
Ethically compliant advanced 3D beamforming for bodypart-scale recognition and tracking through dense massively connected antenna arrays.
- Path 3: Signal processing and learning (WP5):
Ethically compliant high-dimensional tensor processing for the distinction of complex activities and motion from massive-dimensional RF data.