Unified Provisioning Process for Heterogeneous IoT Devices During Mobile Assisted Installation
Inkinen, Iiro (2026)
2026
Tietotekniikan DI-ohjelma - Master's Programme in Information Technology
Informaatioteknologian ja viestinnän tiedekunta - Faculty of Information Technology and Communication Sciences
This publication is copyrighted. You may download, display and print it for Your own personal use. Commercial use is prohibited.
Hyväksymispäivämäärä
2026-06-29
Julkaisun pysyvä osoite on
https://urn.fi/URN:NBN:fi:tuni-202606298026
https://urn.fi/URN:NBN:fi:tuni-202606298026
Tiivistelmä
Different IoT devices have different provisioning needs, yet the installer is best served by a unified installation process. This thesis addresses the tension by extending a mobile-assisted installation flow, attempting to integrate the provisioning of a second sensor into the installation flow originally built for a different one. The goal is an integration that is invisible to the installer.
For the installer, provisioning is just one step of installation. For the device, it’s the only one. During installation, the installer performs multiple integrations during which the device is passive. Only during provisioning, when the device is equipped with the network settings and credentials, it must actively participate in the process. For that reason, the other installation steps aren’t as affected by device heterogeneity, and a unified provisioning may lead to a fully unified installation.
The constructive part of this thesis built a provisioning system for two heterogeneous devices using Design Science Research. The differences between the devices were handled by a single identification step utilizing an NFC chip common to both devices, combined with a modular mobile-application architecture that isolated the device-specific logic. To complete the system, an EDHOC-based provisioning process was designed for the new sensor.
The resulting system met its primary goal of not making changes visible to the user. The device identification step was deemed light in terms of speed and storage, demonstrating the first take-away on how a small common element between otherwise heterogeneous devices can be used to unify the provisioning experience. The provisioning flow for the new sensor also met most of its requirements, falling short only on the expected total provisioning time. The shortcoming was attributed to misplacing the optimization effort based on the expectations from the first sensor, leading to the second takeaway that the provisioning bottlenecks shouldn’t be inherited between devices.
For the installer, provisioning is just one step of installation. For the device, it’s the only one. During installation, the installer performs multiple integrations during which the device is passive. Only during provisioning, when the device is equipped with the network settings and credentials, it must actively participate in the process. For that reason, the other installation steps aren’t as affected by device heterogeneity, and a unified provisioning may lead to a fully unified installation.
The constructive part of this thesis built a provisioning system for two heterogeneous devices using Design Science Research. The differences between the devices were handled by a single identification step utilizing an NFC chip common to both devices, combined with a modular mobile-application architecture that isolated the device-specific logic. To complete the system, an EDHOC-based provisioning process was designed for the new sensor.
The resulting system met its primary goal of not making changes visible to the user. The device identification step was deemed light in terms of speed and storage, demonstrating the first take-away on how a small common element between otherwise heterogeneous devices can be used to unify the provisioning experience. The provisioning flow for the new sensor also met most of its requirements, falling short only on the expected total provisioning time. The shortcoming was attributed to misplacing the optimization effort based on the expectations from the first sensor, leading to the second takeaway that the provisioning bottlenecks shouldn’t be inherited between devices.