Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices

A Master of Science thesis in Computer Engineering by Ragad Moustafa Ahmed entitled, “Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices”, submitted in December 2025. Thesis advisor is Dr. Reham Aburas and thesis co-advisor is Dr. Alex Abraham Aklson. Soft copy...

وصف كامل

محفوظ في:
التفاصيل البيبلوغرافية
المؤلف الرئيسي: Ahmed, Ragad Moustafa (author)
التنسيق: doctoralThesis
منشور في: 2025
الموضوعات:
الوصول للمادة أونلاين:https://hdl.handle.net/11073/33251
الوسوم: إضافة وسم
لا توجد وسوم, كن أول من يضع وسما على هذه التسجيلة!
_version_ 1864513431795662848
author Ahmed, Ragad Moustafa
author_facet Ahmed, Ragad Moustafa
author_role author
dc.contributor.none.fl_str_mv Aburas, Reham
Aklson, Alex
dc.creator.none.fl_str_mv Ahmed, Ragad Moustafa
dc.date.none.fl_str_mv 2025-12
2026-03-23T09:43:12Z
2026-03-23T09:43:12Z
dc.format.none.fl_str_mv application/pdf
dc.identifier.none.fl_str_mv 35.232-2025.67
https://hdl.handle.net/11073/33251
dc.language.none.fl_str_mv en_US
dc.relation.none.fl_str_mv Master of Science in Computer Engineering (MSCoE)
dc.subject.none.fl_str_mv Wearable Sensors
Adversarial Representation Learning
IMU Data
Nymeria Dataset
Gender Inference
Activity Recognition
Egocentric Motion Data
AR Glasses
Data Privacy
dc.title.none.fl_str_mv Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
dc.type.none.fl_str_mv info:eu-repo/semantics/publishedVersion
info:eu-repo/semantics/doctoralThesis
description A Master of Science thesis in Computer Engineering by Ragad Moustafa Ahmed entitled, “Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices”, submitted in December 2025. Thesis advisor is Dr. Reham Aburas and thesis co-advisor is Dr. Alex Abraham Aklson. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form).
format doctoralThesis
id aus_26cd7e7f35547c6a477f8cc31ba3a7bd
identifier_str_mv 35.232-2025.67
language_invalid_str_mv en_US
network_acronym_str aus
network_name_str aus
oai_identifier_str oai:repository.aus.edu:11073/33251
publishDate 2025
repository.mail.fl_str_mv
repository.name.fl_str_mv
repository_id_str
spelling Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable DevicesAhmed, Ragad MoustafaWearable SensorsAdversarial Representation LearningIMU DataNymeria DatasetGender InferenceActivity RecognitionEgocentric Motion DataAR GlassesData PrivacyA Master of Science thesis in Computer Engineering by Ragad Moustafa Ahmed entitled, “Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices”, submitted in December 2025. Thesis advisor is Dr. Reham Aburas and thesis co-advisor is Dr. Alex Abraham Aklson. Soft copy is available (Thesis, Completion Certificate, Approval Signatures, and AUS Archives Consent Form).Wearable motion sensors embedded in devices such as AR glasses, smartwatches, and wristbands capture rich kinematic information that enables accurate activity recognition but simultaneously reveal sensitive demographic traits through unintended inference channels. This thesis investigates the privacy risks inherent in inertial motion data and proposes a privacy-utility preserving representation learning framework based on adversarial training. Using the Nymeria dataset (the world’s largest egocentric motion collection), the study focuses specifically on IMU-derived linear and angular velocity signals from head-mounted and wrist-mounted devices. Although Nymeria has been widely used as a benchmark for foundation models in vision and motion AI, this work presents the first systematic privacy analysis of its sensor modalities. A structured methodological pipeline was developed, including large-scale preprocessing, segmentation, handcrafted feature extraction, baseline modelling, and adversarial representation learning. Baseline results show that the 144-dimensional motion features strongly support script classification but also leak gender information, with a gender inference AUC of 0.889.The proposed Adversarial Representation Learning with Autoencoder (ARA) model suppresses this leakage while maintaining activity-recognition utility: adversarial training reduces gender AUC to 0.5614 (near random guessing) and preserves activity classification utility with an AUC of 0.9738. This thesis represents the first adaptation of an ARA-style adversarial anonymization framework to IMU sensor data from AR glasses and wristbands, demonstrating that adversarial representation learning extends effectively beyond image embeddings to wearable-sensor modalities. Overall, this work establishes an empirical foundation for understanding the privacy vulnerabilities of egocentric motion datasets and provides a practical mechanism for mitigating demographic inference risks in next-generation wearable ecosystems.College of EngineeringDepartment of Computer Science and EngineeringMaster of Science in Computer Engineering (MSCoE)Aburas, RehamAklson, Alex2026-03-23T09:43:12Z2026-03-23T09:43:12Z2025-12info:eu-repo/semantics/publishedVersioninfo:eu-repo/semantics/doctoralThesisapplication/pdf35.232-2025.67https://hdl.handle.net/11073/33251en_USMaster of Science in Computer Engineering (MSCoE)oai:repository.aus.edu:11073/332512026-03-24T05:21:27Z
spellingShingle Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
Ahmed, Ragad Moustafa
Wearable Sensors
Adversarial Representation Learning
IMU Data
Nymeria Dataset
Gender Inference
Activity Recognition
Egocentric Motion Data
AR Glasses
Data Privacy
status_str publishedVersion
title Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
title_full Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
title_fullStr Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
title_full_unstemmed Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
title_short Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
title_sort Developing Privacy Frameworks for Motion Sensor Data in Next-Generation Wearable Devices
topic Wearable Sensors
Adversarial Representation Learning
IMU Data
Nymeria Dataset
Gender Inference
Activity Recognition
Egocentric Motion Data
AR Glasses
Data Privacy
url https://hdl.handle.net/11073/33251