Knowit - Thesis Opportunities 2027

Knowit Connectivity offers five thesis directions covering embedded systems, industrial IoT, Edge AI, and cloud platforms. The projects address challenges in energy efficiency, sensor reliability, device updates, and connected infrastructure. Together with academic and company supervisors, students will refine their chosen topic into research questions, a manageable scope, and evaluation criteria.

The assignment

• Sensor extension: design and evaluate a compact, modular sensor board for low-power embedded and industrial IoT systems. Possible elements include PCB development, power management, sensor interfaces, energy measurements, Zephyr integration, and an initial Cyber Resilience Act assessment. ATEX may inform the design, but certification and formal conformity assessment are excluded.
• Edge AI toolchain: develop a reusable workflow for building, optimizing, deploying, and testing TinyML models on resource-constrained devices. Explore industrial sensor applications such as fault detection and predictive maintenance, with automated deployment and energy-focused benchmarking.
• Cloud platform: build and evaluate a modular reference architecture linking industrial IoT devices to cloud services and managing Edge AI model lifecycles. Address integration, data pipelines, digital device representations, observability, security, portability, and an initial CRA gap assessment.
• Sensor degradation: investigate Edge AI methods for identifying and managing drift, environmental effects, and long-term sensor deterioration. Potential work includes controlled experiments, sensor-health metrics, TinyML models, calibration forecasting, and maintenance support. Accelerated testing provides controlled evidence rather than a substitute for long-term field validation.
• LoRaWAN updates: compare update strategies for large device fleets with differing firmware versions, including delta patches, complete images, multicast, unicast, and patch chains. Evaluate energy consumption, memory requirements, and update duration using ChirpStack, preferably Zephyr, and an existing bootloader such as MCUboot.

Who we are looking for

• Bachelor’s or Master’s student in Computer Science, Embedded Systems Development, or a related discipline.
• Technical experience and interests aligned with the selected project.
• Able to work with an academic supervisor and a Knowit mentor to define and narrow the thesis scope.

What we offer

• A dedicated Knowit mentor providing technical guidance throughout the project.
• Access to the Stockholm office, with attendance for a couple of days per week encouraged.
• Opportunities to meet colleagues and fellow students through after-work activities, workshops, and seminars.

Practical details

Thesis opportunities for 2027 in Stockholm; the exact start date and duration are not specified. Apply through the application link, choose a preferred topic, and attach a CV and current academic transcripts. Students applying as a pair must submit individual applications and name their intended thesis partner. Related project proposals are also welcome, subject to suitable conditions and available mentoring capacity.