Related Experiment Video
Updated: Feb 28, 2026
11:21
Methodology for Establishing a Community-Wide Life Laboratory for Capturing Unobtrusive and Continuous Remote Activity and Health Data
Published on: July 27, 2018
8.9K
Digital Phenotyping via Passive Network Traffic Monitoring: Prospective Observational Study in University Students
Rameen Mahmood1, Annabelle David1, Donghan Hu1
1Department of Electrical and Computer Engineering, Tandon School of Engineering, New York University, 370 Jay St, Brooklyn, US.
JMIR Formative Research
|February 26, 2026
Summary
Passive sensing of encrypted smartphone traffic via VPN is a feasible and acceptable method for digital behavior monitoring. This privacy-preserving technique captures detailed daily activity patterns, offering a scalable tool for health and well-being research.
Area of Science:
- Digital phenotyping and behavioral science
- Mobile health (mHealth) and digital interventions
- Privacy-preserving data collection methods
Background:
- Digital behaviors shape daily life, with disruptions linked to student well-being.
- Existing monitoring methods (wearables, apps, surveys) have limitations in adherence and privacy.
- Passive sensing of network traffic offers a scalable, unobtrusive alternative for capturing smartphone usage.
Purpose of the Study:
- Evaluate the feasibility and acceptability of using VPN-collected encrypted smartphone traffic for digital behavior pattern capture.
- Assess sustained data capture, usability, participant burden, and privacy perceptions.
- Examine if traffic-derived features reveal health-relevant aspects of digital behavior, including timing, intensity, and regularity.
Main Methods:
- Prospective observational study at New York University over two weeks.
- Participants installed WireGuard VPN for passive, encrypted network traffic capture.
- Mixed-methods approach: quantitative (retention, data coverage) and qualitative (interviews, SUS, NASA-TLX) for feasibility and acceptability.
Main Results:
- 29 students formed the analytic cohort, with 71% overall retention and 93% contributing at least five days of data.
- High usability (SUS=78) and low perceived workload (NASA-TLX minimal) reported by participants.
- Exploratory analysis showed traffic features reflected daily rhythms and lifestyle patterns (e.g., gaming, food delivery use).
Conclusions:
- VPN-based monitoring of encrypted smartphone traffic is feasible, acceptable, and enables sustained, privacy-preserving data collection.
- This method shows promise for scalable, device-agnostic digital phenotyping of behavioral rhythms.
- Further validation could enhance its utility in studying everyday health and well-being.
