PP-IDE Behavioural Device Fingerprinting Dataset for Privacy-Preserving Identity Establishment
Abstract
This dataset provides a processed behavioural device-fingerprinting representation designed to support research on privacy-preserving identity establishment and continuous device authentication. The dataset is derived from controlled execution-response measurements collected from 11 Raspberry Pi devices operating under fixed experimental conditions. Measurements originate from 3 Raspberry Pi 3 Model B+ devices and 8 Raspberry Pi 4 Model B devices. Data collection was conducted in a controlled headless execution environment with fixed CPU frequency configuration, CPU core isolation, and reduced scheduling interference to ensure reproducibility of behavioural response signals. Each device was evaluated across 4 CPU cores and multiple reboot cycles. The dataset contains execution-dependent behavioural signals, including: Temperature-dependent execution responses GPU-related performance counter behaviour CPU hash execution response signals Pseudo-random execution response signals True random number generation: behavioural signals A comprehensive feature engineering pipeline was applied, incorporating: Temporal rolling statistical features Exponential moving dynamics Signed behavioural interaction features Robust statistical transformations Behavioural deviation modelling Representation-learning-oriented feature construction Unlike the CD2A dataset, which utilises a limited handcrafted feature representation focused primarily on QPU frequency behaviour, the PP-IDE dataset…
Read the paper
Related studies
- Active Learning during Federated Learning for Object Detection
- Fairness-aware federated learning
- Decentralized Federated Learning
- Building a Personally Identifiable Information Recognizer in a Privacy Preserved Manner Using Automated Annotation and Federated Learning
- Preservation of Patient Level Privacy: Federated Classification and Calibration Models
- Efficient explainable federated learning
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.