Smarter Firmware Updates for Large IoT Device Fleets
How do you update thousands of connected devices without wasting bandwidth, battery power, or memory?
Description
Imagine managing thousands of IoT devices deployed across industrial environments. Some devices are running the latest firmware, others have missed previous updates, and many operate with limited battery capacity and wireless bandwidth.
Updating all these devices efficiently is a significant engineering challenge.
At Knowit Connectivity, we are offering a thesis opportunity focused on developing a smarter approach to firmware updates for large LoRaWAN device fleets.
About the Project
The project investigates how delta firmware updates can reduce the resources required to update connected devices.
Instead of transferring an entire firmware image, a delta update sends only the differences between two firmware versions.
However, this becomes more complicated when devices within the same fleet are running different versions of the software.
The goal is to develop a system that identifies each device's firmware version, groups devices appropriately, and selects an efficient update strategy.
You will explore how multicast, unicast, delta patches, and full firmware images can be combined to minimize energy consumption, memory requirements, and update overhead.
The work will involve LoRaWAN communication, embedded software, Zephyr OS, and integration with the ChirpStack network server.
What You Will Explore
Your work may include:
Investigating why large IoT fleets end up running different firmware versions.
Evaluating existing delta-update tools suitable for microcontrollers.
Comparing patch sizes, memory usage, flash wear, and energy consumption.
Developing methods for grouping devices based on installed firmware versions.
Designing logic for selecting between multicast, unicast, delta patches, and full firmware updates.
Integrating update management with ChirpStack.
Implementing delta-update functionality on embedded devices, preferably using Zephyr OS.
Measuring update time, radio listening time, RAM usage, flash operations, and energy consumption.
Evaluating the solution using representative fleet scenarios and real hardware measurements.
Who We're Looking For
This thesis could be suitable for students studying Embedded Systems, Computer Engineering, Computer Science, Electrical Engineering, Telecommunications, or a related field.
We believe you have an interest in:
Embedded software development.
C programming and microcontrollers.
Wireless communication and IoT.
LoRaWAN and low-power networks.
Firmware architecture and bootloaders.
Zephyr OS or similar embedded environments.
Performance optimization and resource-constrained systems.
An interest in understanding how software decisions affect real hardware performance will be particularly valuable.
Expected Outcomes
The thesis aims to deliver a working firmware-update solution capable of managing devices running different firmware versions.
The solution should include device grouping, update-strategy selection, ChirpStack integration, and embedded delta-update functionality.
You will evaluate how different strategies affect energy consumption, memory requirements, flash wear, and update performance.
The results should provide practical guidance for managing firmware updates in larger industrial IoT deployments.
The final scope and research questions will be defined together with your supervisors.
Interested in This Thesis?
Interested in embedded software, wireless communication, and solving practical challenges in large-scale IoT systems?
Apply through our thesis application page and indicate your interest in "Delta Firmware Updates for LoRaWAN Fleets".