Designing a Secure and Resilient Distributed Smartphone Participant Data Collection System
Foad Namjoo, Neng Wan, Devan Mallory, Yuyi Chang, Nithin Sugavanam, Long Yin Lee, Ning Xiong, Emre Ertin, Jeff M. Phillips · 2025
Read Designing a Secure and Resilient Distributed Smartphone Participant Data Collection System by Foad Namjoo, Neng Wan, Devan Mallory, Yuyi Chang, Nithin Sugavanam, Long Yin Lee, Ning Xiong, Emre Ertin, Jeff M. Phillips (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2510-19938.txt.
Introduction
Foad Namjoo and eight co-authors describe MotionPI, a smartphone system for collecting behavioral and health data from study participants securely and with little effort on their part. Real health studies need continuous data under messy conditions: dying batteries, weak cellular signals, and unsupervised users. MotionPI gathers passive sensor data, including location and wristband motion, and prompts short in-the-moment surveys, triggered randomly or by activity.
Data are encrypted in transmission and storage, kept both on the phone and on a secure cloud server. A Bluetooth wristband buffers raw data and sends motion summaries and trigger events to the phone. The design aims at practical, scalable data collection for cyber-physical health research without burdening participants.
Catalog record
| CREATOR | Foad Namjoo, Neng Wan, Devan Mallory, Yuyi Chang, Nithin Sugavanam, Long Yin Lee, Ning Xiong, Emre Ertin, Jeff M. Phillips |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 23 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2510-19938
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Designing a Secure and Resilient Distributed Smartphone Participant Data Collection System Thanks: This work was supported by the National Cancer Institute of the National Institutes of Health under award number R37CA276365. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. Foad Namjoo, Neng Wan, Devan Mallory, Yuyi Chang, Nithin Sugavanam, Long Yin Lee, Ning Xiong, Emre Ertin, Jeff M. Phillips Abstract Real-world health studies require continuous and secure data collection from mobile and wearable devices. We introduce MotionPI , a smartphone-based system designed to collect behavioral and health data through sensors and surveys with minimal interaction from participants. The system integrates passive data collection (such as GPS and wristband motion data) with Ecological Momentary Assessment (EMA) surveys, which can be triggered randomly or based on physical activity. MotionPI is designed to work under real-life constraints, including limited battery life, weak or intermittent cellular connection, and minimal user supervision. It stores data both locally and on a secure cloud server, with encrypted transmission and storage. It integrates through Bluetooth Low Energy (BLE) into wristband devices that store raw data and communicate motion summaries and trigger events. MotionPI demonstrates a practical solution for secure and scalable mobile data collection in cyber-physical health studies. 1 Kahlert School of Computing, University of Utah [email protected], [email protected] 2 School of Environment, Society & Sustainability, University of Utah [email protected], [email protected], [email protected] 3 Department of Electrical and Computer Engineering, The Ohio State University [email protected], [email protected], [email protected], [email protected] Keywords— System Design; Cyber-Physical Systems; Mobile Health; Wearable Sensors 1 Introduction Understanding and improving personal health behavior is a major challenge. Although advances in smart devices make it possible to capture fine-grained medical, personal, and movement data about individuals, doing so as part of a medical study still presents numerous challenges. The resulting system must be inexpensive, automated, easy-to-use, reliable, and protect private patient data. This requires a carefully designed secure smartphone system. General personalized health studies require multiple components. First, movement is a key indicator of fitness and activity, and its relationship to health outcomes is an important factor. As such, smart systems need to collect various high-frequency statistics, such as GPS location and tracking of wrist movement and heart rate through peripheral smart devices. Beyond just tracking, many health studies require interventions , where the smart app detects certain activities and prompts the study patient to perform a task. This task may also collect data like a survey. To achieve these goals, the smartphone interface should be minimal and seamless to ensure that it is used and that participants only use the intended features. It should be implemented efficiently with only the necessary Bluetooth connections to peripherals and other storage costs, so the batteries of the equipment can last a full day without recharging. Figure 1: Diagram of the MotionPI Ecosystem. The App plays a central role in coordinating the data collection from peripherals, high-frequency sensing, and EMAs, and its secure and distributed storage. When dealing with private patient health information, it must be secure. The equipment provided to study participants should not have residual data kept on the device of previous participants and if it is there, it should be encrypted. The storage method should also be resilient and verifiably consistent. These studies need many participants, and it is a complex process to recruit them and convince them to actively participate in the procedures. If data is lost, it is a waste of time and expense. Our solution. We provide a smartphone system that stores data in multiple ways, both in the smartphone storage (with encryption) and in a secure cloud environment. This setup enables distributed, simultaneous, and secure uploads directly into a NoSQL (MongoDB) database. This structure, shown in Figure 1 , generates many advantages: (1) It is easy to monitor the data collection process and quality, (2) It enforces access control so that only authorized persons and all data analysts part of the project can access the data at any time, and (3) It enables verification that all the collected data has been successfully uploaded by comparing them with the locally stored files. To accomplish this, our proposed system, MotionPI, needs to make several critical design decisions. These include how the data are stored (both format and server environment), how to implement a secure container deployment, how to ensure data is eventually sent to the cloud when the smartphone’s connectivity is sporadic, and it must ensure efficiency so it will not drain the smartphone battery too quickly. We document how we solve these issues in this report. Our system is currently being actively deployed in a large NIH-funded study. 2 Related Work Several mobile health data collection platforms have been developed to support behavioral and physiological studies in the real world. Among the most widely used are mCerebrum [ 5 ] , Beiwe [ 8 ] , and RADAR-base [ 9 ] . Although mCerebrum is a widely used mobile health platform that supports sensor data collection and digital biomarkers, we encountered compatibility issues when trying to update and run the software on newer versions of Android. The mCerebrum framework offers extensive sensing capabilities and supports high-frequency data collection, but its configuration requires technical expertise and often relies on continuous network connectivity. Beiwe provides strong privacy guarantees by encrypting data before upload, but is designed primarily for centralized studies and lacks support for edge-level computation or offline-triggered surveys. RADAR-base uses wearable devices and smartphones to stream data in real time, but its architecture assumes stable connectivity. Other platforms, such as AWARE [ 4 ] , allow sensor-based logging, but offer limited support for secure local storage or adaptive EMA logic. In contrast to these systems, our proposed platform, MotionPI , is specifically designed for deployments with limited network availability, minimal participant interaction, and high privacy requirements. MotionPI enables encrypted local storage, robust background uploading to a secure cloud server, and EMA scheduling logic triggered by both time and sensor thresholds. MotionPI combines a dual storage architecture (encrypted local and cloud redundancy) with timed EMA triggers —including BLE-based physical-activity detection (based on ENMO [ 1 ] )—and device-scoped authentication for offline operation that tolerates intermittent connectivity; see Table 1 . The remainder is necessary engineering integration (phone app + wristbands + backend + ops) that makes the system usable in real deployments while preserving confidentiality and availability under weak networks. Table 1: Comparison of mobile data collection platforms. MotionPI supports offline-triggered surveys, BLE-based activity detection, and dual storage (local + cloud). Platform Timed EMA trigger BLE activity trigger Dual storage Offline sign-in mCerebrum [ 5 ] yes yes no no Beiwe [ 8 ] yes no yes no RADAR-base [ 9 ] yes no no no AWARE [ 4 ] yes no yes via config URL MotionPI yes yes yes yes 3 The Smartphone App The MotionPI mobile application is a secure, resilient, and highly modular Flutter-based system developed to collect behavioral and physiological data in real-world health studies. It integrates passive (e.g., GPS and motion) and active assessments (surveys) through sensors and Bluetooth-connected wristbands. The app operates continuously with minimal user interaction and handles limited connectivity and low-power environments. It automatically operates