<!DOCTYPE article PUBLIC "-//NLM//DTD JATS (Z39.96) Journal Archiving and Interchange DTD v1.0 20120330//EN" "JATS-archivearticle1.dtd">
<article xmlns:xlink="http://www.w3.org/1999/xlink">
  <front>
    <journal-meta />
    <article-meta>
      <title-group>
        <article-title>Leveraging Zephyr RTOS for modern embedded systems A perspective from education and research</article-title>
      </title-group>
      <contrib-group>
        <contrib contrib-type="author">
          <string-name>Sven Grunwald</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <contrib contrib-type="author">
          <string-name>Kevin Krebs</string-name>
          <xref ref-type="aff" rid="aff0">0</xref>
        </contrib>
        <aff id="aff0">
          <label>0</label>
          <institution>TUD Dresden University of Technology, Chair of Information Technology for Traffic Systems (ITVS)</institution>
          ,
          <addr-line>01069 Dresden</addr-line>
          <country country="DE">Germany</country>
        </aff>
      </contrib-group>
      <pub-date>
        <year>2026</year>
      </pub-date>
      <abstract>
        <p>Due to its strong community support and portability, the Zephyr Real-Time Operating System (RTOS) is gaining increasing popularity in both academic and industrial environments. This paper discusses the integration of Zephyr RTOS into research projects with a particular focus on its application at the Chair of Information Technology for Transport Systems (ITVS) at TU Dresden. We highlight how Zephyr facilitates cutting-edge research in networked embedded systems and serves as a foundation for practical instruction in IoT and telematics. Additionally, we examine the educational benefits but also the technical challenges.</p>
      </abstract>
      <kwd-group>
        <kwd>eol&gt;Zephyr RTOS</kwd>
        <kwd>education</kwd>
        <kwd>lectures</kwd>
        <kwd>classroom courses1</kwd>
      </kwd-group>
    </article-meta>
  </front>
  <body>
    <sec id="sec-1">
      <title>1. Introduction</title>
    </sec>
    <sec id="sec-2">
      <title>2. Rationale for using Zephyr RTOS in education and research</title>
      <p>When talking about embedded devices there are usually two main approaches: classic bare-metal
programming approaches and RTOS-based approaches. Easily someone would say that the
baremetal approach provides direct control over hardware resulting in maximum efficiency in
1† These authors contributed equally.</p>
      <p>sven.grunwald@tu-dresden.de(S. Grunwald); Kevin.Krebs@tu-dresden.de (K.Krebs)</p>
      <p>© 2025 Copyright for this paper by its authors. Use permitted under Creative Commons License Attribution 4.0 International (CC BY 4.0).
combination with minimal delay and a small memory footprint that is ideal for simple or high
resource-world systems. Which may be correct, but this means an increase in the complexity of
development, lack of portability and the absence of underlying multitasking is difficult to manage
concurrent functions or scale the system. On the other hand, using an RTOS (real-time operating
system) provides support for structured work management, hardware abstraction and multitasking
that greatly simplifies development for complex systems and increases scalability and maintenance
but there’s a certain learning curve and the need to understand some basic concepts in programming
and RTOS-driven software development. Central to our teaching and research is the Zephyr RTOS.
It plays an important role in enabling practical and innovative work with embedded systems since
2020 within the classroom as well as within research projects at the chair at TU-Dresden.</p>
      <p>
        Zephyr’s modular architecture allows students and researchers to tailor the operating system to
specific project needs, gathering a deep understanding of embedded system design by selecting only
the necessary components and protocols [
        <xref ref-type="bibr" rid="ref1">1</xref>
        ]. Due to the wide range of processor architecture support
and hardware platforms it enables seamless transitions between various devices and projects,
preparing students for various engineering challenges they will cope with [
        <xref ref-type="bibr" rid="ref2">2</xref>
        ]. Zephyrs real-time
capabilities ensure resistant behavior and low delays which are important for time-sensitive
applications in telematics and IoT [
        <xref ref-type="bibr" rid="ref4">4</xref>
        ]. Also integrated features such as memory protection and
cryptographic libraries enable the development of strong and secure applications.
      </p>
      <p>
        This is becoming more and more important in a globally connected world and is underlining the
importance of cybersecurity in modern IoT systems [
        <xref ref-type="bibr" rid="ref5">5</xref>
        ]. The open source of Zephyr provides
widespread documentation, development equipment and learning resources which accelerate the
learning process, promotes experiments and supports collaborative research. With Zephyrs’
extensive adoption in the IoT industry our students get very relevant skills that transfer directly to
the professional environment effectively removing the distance between educational education and
industry requirements. Once students and developers become familiar with the configuration-driven
approach they will find that enabling hardware features such configuring GPIOs and switching on
LEDs is like activating more sophisticated subsystems such as wireless connectivity or networking.
This stability means that whether you are working with basic devices / functionalities or advanced
connectivity stacks the process usually involves enabling appropriate configuration options and
adjusting the device tree [
        <xref ref-type="bibr" rid="ref6">6</xref>
        ]. As a result, the learning state invested in understanding Zephyr’s
configuration model pays off across the entire platform making it easier to scale from simple projects
to more sophisticated and feature-rich applications. This cohesive design also reduces the potential
for errors and accelerates the transition from prototype to production, highlighting the robustness
and versatility of the Zephyr ecosystem. Thats a welcome relief for anyone who’s ever found
themselves debugging cryptic faults or stuck in a hard fault handler. With Zephyr the path from idea
to implementation is more predictable and far less daunting.
      </p>
      <sec id="sec-2-1">
        <title>2.1. Curriculum integration</title>
        <p>At TU Dresden, we offer a variety of courses for both undergraduate and graduate students that
use the Zephyr Real-Time Operating System (RTOS) to connect theory with hands-on practice. Our
students come from many different fields including electrical engineering, computer science,
aeronautics, rail transport and traffic telematics. This diversity brings some challenges but also great
opportunities. Notably, almost none of the students have prior experience working with
microcontroller boards and many lack a formal background in communications engineering. Yet,
understanding connectivity technologies is becoming increasingly important across all areas of
technology. To support this, we have developed a series of “crash courses” that run alongside our
main lectures. These courses focus on key concepts in embedded systems, networking and wireless
communications while giving students practical experience using Zephyr as a teaching tool. This
approach helps ensure that all students, regardless of their background or previous hands-on
experience gain the essential knowledge and skills needed to work effectively in today’s connected
world.</p>
        <p>Lectures at ITVS</p>
        <p>Zephyr building blocks within lectures
Traffic Sensor Technology</p>
        <p>Digital Signal Processing</p>
        <p>Traffic Telematics Networks
Theory and Technology of Information Systems
7th semester
6th semester
6th semester
5th semester
Audience</p>
        <p>Research projects
Electrical Engineering</p>
        <p>Traffic Telematics
Railway Systems and Public Transport</p>
        <p>Air Transportation and Logistics</p>
        <p>External peripherals e.g. sensors</p>
        <p>CMSIS DSP / Algorithms
Networking / Wireless connectivity
Basic usage / RF-Basics antennas</p>
        <p>Custom hardware e.g. nRF91
Custom traffic sciences related
applications e.g. IQ-sampling
Algorithmic / signal processing tasks</p>
        <p>
          To provide a clearer picture of how these concepts are integrated into our curriculum, the
following section outlines some of the key lectures and their focus areas. Within lectures such as
Theory and Technology of Information Systems we are establishing a strong foundation by teaching
the fundamentals of radio frequency technology, antenna principles and propagation effects essential
wireless basics for students without a background in communications engineering [
          <xref ref-type="bibr" rid="ref9">9</xref>
          ]. This lecture
focuses on more low-level hardware near topics. Those skills are important because the underlying
hardware in any communication system needs to be understood to make decisions about its
suitability in a future application. This theoretical groundwork is seamlessly connected to practical
experience: students leverage Zephyr RTOS to implement and analyze real-world scenarios such as
state-of-the-art network protocol operations gaining direct insight into how data propagates through
complex systems. In complementary courses like Traffic Sensor Technology, Digital Signal
Processing and Traffic Telematics Networks our students further explore the practicalities of modern
connectivity [
          <xref ref-type="bibr" rid="ref9">9</xref>
          ]. They work with widely used protocols (e.g., CoAP, UDP and TCP) as well as
emerging IoT standards such as Thread, Matter and UWB for wireless sensor nodes. Another
important topic in the curriculum when dealing with networked devices is cybersecurity. By teaching
the core concepts of cryptography with the help of Alice and Bob, the students can quickly delve
into more specific areas in the field. Learning how methods can be applied to grant access to data
only to specific users (authorization and authentication) and means to check whether information
was altered or tampered with (integrity). These principles are not only introduced in lectures but
are also explored in laboratory exercises where students design and develop sensor nodes, gateways
and smart mobility testbeds in combination with RF-measurement equipment. Through these
projects, they gain experience with embedded systems concepts like RF-performance, data
throughput, real-time scheduling, power management, multi-threading and device driver
development, e.g. accelerometer sensitivity range adjustments for a given application. Our
curriculum tackles the following key areas:
•
•
•
        </p>
        <p>Fundamentals in RF-technology, antenna principles and propagation effects
The ISO/OSI layer model and practical networking, reinforced through hands-on work
with protocols like MQTT / MQTT-SN, CoAP, UDP and TCP
Introduction and application of modern IoT connectivity standards such as Thread,
Matter, UWB preparing students to engage with real-world smart device ecosystems
•
•</p>
        <p>Fundamentals in Cybersecurity and best-practices for data security in IoT applications
and why it is necessary
Core embedded systems topics e.g. power, constraints and memory management
Design and implementation of sensor nodes and gateways for smart mobility and traffic
telematics, effectively bridging theoretical concepts with industry-relevant topics
By integrating Zephyr into the curriculum, we ensure that students from diverse academic
backgrounds acquire the connectivity and embedded systems skills that are now fundamental across
all modern engineering and technology fields. Leveraging Zephyr’s extensive library of sample
applications and board support packages enables students to focus on high-level system design,
debugging and application development fast. This approach accelerates the learning process but also
mirrors best practices in industry where rapid prototyping and cross-disciplinary collaboration are
the norm and kind of expected. All in all the curriculum is designed to make complex technical topics
accessible and relevant to students from all engineering backgrounds ensuring that they acquire the
essential skills needed to thrive in a world where connectivity is ubiquitous and foundational to
innovation in every technical field. Through this holistic and application-oriented integration of
Zephyr we empower our students to understand but also design and troubleshooting modern
connected systems.</p>
      </sec>
      <sec id="sec-2-2">
        <title>2.2. Case Study: research project on real-time data acquisition</title>
        <p>At TU Dresden a recent research project showcased how powerful Zephyr RTOS can be for
realtime data acquisition in traffic monitoring. The team worked with a custom hardware platform based
on the Nordic Semiconductors® nRF9160 cellular IoT module. Due to the modular software design
and the available configuration tools the researchers were able to build a robust multi-layer network
stack and integrate secure cloud connectivity via MQTT. The components of the setup are shown in
figure 3. The protocol MQTT was chosen to ensure flexibility, scalability and loose coupling to have
freedom in the mode of operation on both sides. The development of a live web dashboard for data
visualization was part of the outcome as shown in figure 4.
moved from theoretical models such as real-time scheduling and secure communication protocols
into real-world implementation. The team could work “close to the metal” when needed, optimizing
low-level peripherals (e.g. ADC) and power consumption, but also take advantage of abstractions for
rapid development which is usually the case when dealing with research projects and tight timelines
especially in a university context.</p>
        <p>A clear illustration of Zephyr’s strengths emerged when the team needed to implement a custom
low-power sleep mode that dynamically adjusted based on physical movement and incoming sensor
data rates and network activity. Using Zephyr’s flexible power management APIs and real-time
kernel features they could fine-tune the device’s energy consumption without sacrificing
responsiveness. For instance, the system was able to wake instantly to process high-priority traffic
events and then return to deep sleep ensuring to minimize power consumption. This was an essential
requirement for cellular IoT deployments within the remote traffic monitoring context because the
test carrier was quite sensitive to quiescent current. This level of control in combination with
Zephyr’s support for seamless integration of custom analytics modules enabled the team to process
and filter large volumes of sensor data locally e.g. sending only relevant information to the cloud. As
a result they achieved both efficient energy usage and reduced the costs of the cellular data
demonstrating how Zephyr empowers developers to create optimized and field-ready solutions.</p>
        <p>Zephyr based
connectivity system
The outcomes of the project were not confined to research activities alone. They were
systematically integrated into our teaching framework, facilitating the adoption of the Zephyr RTOS
and nRF9160 platforms in our curriculum. Challenges encountered and solutions developed during
the research phase were translated into new laboratory exercises and project modules that closely
reflect real-world engineering scenarios. This approach provides students with concrete examples
that bridge theoretical instruction and practical application, thereby demystifying complex
embedded systems concepts through experiential learning. For instance, students engage directly
with the same hardware and firmware utilized in our research allowing them to observe firsthand
the implementation of advanced topics such as power management and secure IoT communication
including the practical application of cryptographic techniques. Furthermore, students experiment
with various cryptographic algorithms gaining insight into their impact on system resources such as
memory utilization. This hands-on exposure equips students with the skills and understanding
necessary to address contemporary challenges in embedded and connected systems. A more complex
and therefore challenging task for the participants is setting up the Zephyr environment to work
with network protocols like MQTT and MQTT-SN making sure everything runs well on low-power
devices. One important assignment is to keep the system reliable even when things go wrong, like
losing the network or server issues arise. Through this practical experience, students not only tackle
the tough topics that come with modern RTOS development but also gain confidence to play around
with advanced features like remote diagnostics and cloud data visualization (Figure 5). They learn to
fix real problems, adapt to changing needs and understand why solid design matters. This hands-on
approach gives them a deeper grasp of embedded systems and gets them ready to face challenges in
the industry with practical skills and creative thinking.
By engaging with industry-standard tools and development workflows students gain valuable skills
that are directly transferable to professional settings. The integration of Zephyr and the nRF9160
platform into the curriculum has inspired many students to pursue their own projects and even take
a closer look into open-source initiatives. Over time this work has also helped strengthen our
partnerships with leading industry players like Nordic Semiconductor, ST Microelectronics,
Memfault e.g. guest talks from an industry perspective. These collaborations have opened doors for
students to work with cutting-edge hardware and software such as Nordic’s nRF Connect SDK,
stateof-the-art hardware platforms like nRF53/54/91, Memfault’s advanced device monitoring tools and
even STMicroelectronics™ MEMS sensors featuring machine learning core technology. Not only do
students gain hands-on experience with the same technologies used by professionals but they also
have opportunities to connect with industry mentors and build networks early. This is especially
valuable in the Dresden region where the embedded sector is thriving. Through workshops, guest
lectures and joint projects our students are exposed to real-world challenges and current industry
practices further enhancing their readiness for the workforce. As more companies adopt Zephyr this
practical exposure and industry engagement give our students a real advantage in their studies and
as they launch their professional journeys. This synergy between academia and industry ensures
that our graduates are well-prepared to contribute meaningfully to the rapidly evolving field of
embedded systems.</p>
      </sec>
    </sec>
    <sec id="sec-3">
      <title>3. Lessons learned and best practices</title>
      <p>The inclusion of Zephyr RTOS in our embedded systems class has been a driving force in
advancing the adoption of teaching ideas in real-time and embedded software. The scalability of
Zephyr's design, with the aid of a microkernel that is supported by many hardware platforms, is an
environment relevant to the industry for students. Early engagement with Zephyr allows students
to master essential RTOS design patterns, including preemptive multitasking, priority-based
scheduling and inter-thread communication.</p>
      <sec id="sec-3-1">
        <title>3.1. Transformative impact and collaborative learning with Zephyr RTOS</title>
        <p>
          The adoption of Zephyr RTOS within embedded systems education has led to significant changes
in both instructional content and pedagogical methods. Curricula increasingly combine theoretical
lectures with practical laboratory sessions enabling comprehensive coverage of essential RTOS
principles including kernel initialization context switching, interrupt management and various
memory management techniques such as heap, stack and memory pool utilization. Laboratory-based
activities reinforce these foundational concepts by requiring participants to deploy Zephyr on
diverse hardware platforms modify device trees for hardware abstraction and interact with
subsystems such as file systems, network stacks and peripheral drivers. This emphasis on
experiential learning facilitates the transition from theoretical understanding to practical
competence. Application-oriented tasks such as the implementation of sensor fusion algorithms or
the development of Bluetooth Low Energy (BLE) solutions serve to bridge the divide between
academic instruction and industry practices. A central factor in this educational evolution is the
open-source nature of the Zephyr ecosystem. The collaborative environment fosters exploration of
kernel source code contributions to bug fixes and feature enhancements and active engagement with
a global developer community. Such exposure demystifies the software development process and
encourages the adoption of professional practices including code review, version control and
continuous integration. By shifting the focus toward application logic and overall system design
Zephyr’s abstraction mechanisms make it possible to engage with sophisticated subjects including
Bluetooth LE channel sounding without requiring exhaustive familiarity with low-level hardware
details we learned recently in a student thesis we supervised at the chair [
          <xref ref-type="bibr" rid="ref10">10</xref>
          ].
        </p>
      </sec>
      <sec id="sec-3-2">
        <title>3.2. Addressing the learning curve</title>
        <p>Although it has its advantages, Zephyr does have a non-negligible learning curve, particularly for
users from bare-metal or proprietary embedded development backgrounds. Key hurdles to overcome
are understanding Zephyr's build system (west and CMake), the Kconfig configuration language and
adapting to the device tree hardware description paradigm. These are elegant concepts but can be
mysterious to the uninitiated and require a shift in thinking from imperative register manipulation
to declarative hardware configuration. To assist in mitigating these concerns, we are and continue
to make the following best practices:
•
•
•
•
•
•</p>
        <p>Begin with foundational Zephyr topics such as the basic application structure and build
workflow before advancing to more complex features like multi-threading and
interprocess communication
Integrate comprehensive, hands-on tutorials that walk students through essential tasks
such as configuring hardware with device tree overlays and customizing system behavior
with Kconfig using practical examples
Structure coursework so that each project builds upon the previous one, gradually
introducing new Zephyr subsystems and concepts to reinforce learning and promote a
sense of achievement
Encourage teamwork through code reviews, group debugging sessions and peer-to-peer
mentoring encouraging students to share insights and troubleshooting together
Utilize regular check-ins, surveys and reflective exercises to identify learning bottlenecks
and adapt instructional methods accordingly
Empirical results collected in the form of pre- and post-course survey indicate a
substantial rise in students' self-rated confidence and skill level in RTOS concepts after
completing Zephyr-based modules. Performance indicators such as lab completion rates
and project success rates also affirm these results.</p>
      </sec>
      <sec id="sec-3-3">
        <title>3.3. Paradigm shift: from register-level programming to system-level design</title>
        <p>One significant result of bringing Zephyr into the classroom is the shift it encourages in how
students approach embedded systems. Instead of focusing on programming at the register level
students begin to think in terms of system-level design and abstraction. In many traditional
embedded systems courses the emphasis is placed on directly manipulating peripheral registers.
While this method has educational value it does not always align with the way embedded software
is developed in industry today especially when talking connected embedded systems. Zephyr
changes this by providing device drivers and standardized application programming interfaces. This
results in highly portable code that can run on a variety of architectures e.g. ARM Cortex-M or
RISCV. This abstraction makes it easier for students to direct their efforts toward building systems that
are scalable and easier to maintain. For example, instead of writing code that is specific to a particular
vendor just to initialize the UART. Students can use Zephyr’s common interface for serial
communication. This optimizes the development process and introduces important concepts such as
platform independence as well as modularity and reuseable code. These are qualities that are highly
valued in the current embedded systems job market. It is important to acknowledge that Zephyr’s
abstraction does not always cover every possible use case with complete consistency. There are
situations where applications need access to advanced or highly specialized hardware features and
in these cases the generic interfaces may not provide all the necessary functionality. For example,
Nordic Semiconductors® nRF devices using features like the programmable peripheral interconnect
(PPI) for advanced analog-to-digital conversion often require developers to work directly with the
vendor’s software development kit stepping outside of Zephyr’s abstraction. The same is true for the
Raspberry Pi Pico where the basic analog-to-digital converter can be accessed through Zephyr but
making use of advanced features such as direct memory access, round-robin sampling or detailed
configuration options typically means turning to the Pico SDK or manipulating hardware registers
directly. These examples illustrate a challenge in the design of portable system code. There is often
a trade-off between generating code that’s portable across different hardware platforms and to be
able to access the full range of features that a specific device offers. While Zephyr is designed to
promote portability and modularity developers sometimes need to fall back on hardware-specific
code to achieve the best performance or to use unique features. This reality highlights why it remains
important for students to understand both the higher-level concepts of operating systems and the
lower-level details of hardware programming especially when working on advanced or
performancesensitive embedded applications. By explicitly addressing these limitations in the curriculum
students gain a realistic understanding of both the strengths and boundaries of RTOS-based
development. They learn not only to appreciate the benefits of system-level abstraction but also to
recognize scenarios where direct hardware access remains essential for leveraging the full potential
of modern microcontrollers.</p>
      </sec>
      <sec id="sec-3-4">
        <title>3.4. Quantitative evaluation of learning outcomes</title>
        <p>The integration of Zephyr RTOS into our embedded systems courses has produced measurable
improvements in student performance, confidence and project success. Oral surveys before and after
implementation indicate that students' self-reported confidence in developing embedded applications
increased significantly. From around 40% confidence initially to 75% upon completing Zephyr-based
courses illustrating how hands-on exposure to a modular, industry-focused RTOS bridges the
knowledge-practice gap. Project and laboratory success rates also consolidate this trend. Successful
project submission went up from 65% in previous cohorts to 85% after Zephyr was integrated, with
significant improvement in applications requiring lots of resources, including real-time processing
of data and power management. Additionally, lab exercise debugging time for minor firmware issues
decreased by approximately 50%, from 30 minutes average time to 15 minutes. This indicates that
students rapidly gain debugging tool proficiency and system knowledge for less complex
programming flaws. These gains are complemented by more instances of using optional advanced
projects, such as building secure IoT communication modules or low-power sensor networks, which
leverage Zephyr's on-board security libraries and power management APIs. In combination, these
quantitative results demonstrate that Zephyr RTOS increases technical expertise and builds a deeper
knowledge of system-level design principles critical to modern embedded development. In the future,
longitudinal studies will attempt to measure how these initial successes impact students' transition
into industry roles, with the hope that the experience and exposure accrued through Zephyr will
mean greater employability and readiness for future needs within the field of embedded systems.</p>
      </sec>
      <sec id="sec-3-5">
        <title>3.5. Debugging Challenges and the need for improved tooling</title>
        <p>The most longstanding problem encountered is most likely debugging within the Zephyr
platform. Tightly integrated but proprietary IDEs such as KEIL® uVision® and IAR Embedded
Workbench®, for instance, offer a more straightforward setup process and experience. Since Zephyr
relies on open-source and free tools like GDB, OpenOCD and SEGGER Ozone. This setup procedure
can be intimidating for both students and instructors alike, especially for those new to embedded
systems development. Moreover, the use of tools occasionally results in unpredictable behavior
during debugging or programming sessions. Issues such as devices failing to program without a
visible feedback or unexpected deadlocks at startup are not uncommon and can detract from the
overall learning experience. To address these debugging challenges there is a need for better tooling
and integrated development environments that simplify setup and troubleshooting. Developing or
adopting Zephyr-specific IDEs with enhanced features such as real-time thread visualization and
integrated peripheral monitoring could significantly improve the user experience. Obviously
debugging multi-threaded, real-time programs introduces additional and challenging complications.
Deadlocks, priority inversion and race conditions are intrinsically difficult to identify, as context
switching and task scheduling events are not always obvious in the source code or easily traced using
traditional debug tools. Current solutions are mainly for single-threaded static platforms and do not
adequately deal with the dynamic behavior of RTOS-based applications. To address these challenges,
we recommend the following from a teaching perspective:
•
•
•
•</p>
        <p>Creation or adoption of a graphical interface supporting real-time thread visualization,
event tracing and integrated peripheral monitoring
Comprehensive instructional materials and video guides for setup and use of debugging
tools emphasizing common challenges and advanced RTOS debugging methods
Incorporation of advanced analysis solutions e.g. Percepio Tracealyzer® or SEGGER
SystemView to provide accessible, real-time insight into thread execution, system events
and resource usage in a more streamlined and easy to follow way
Collection and analysis of data on debugging outcomes, resolution times and user
feedback to continuously improve curriculum effectiveness</p>
      </sec>
      <sec id="sec-3-6">
        <title>3.6. Recommendations for future improvement and value additions</title>
        <p>To further enhance the educational value of Zephyr in the classroom and contribute to the
broader community the following initiatives could be useful:
•
•
•
•</p>
        <p>Systematically gather and analyze quantitative and qualitative data on student
performance and learning outcomes
Contribute modular, open-source teaching materials including lab manuals, annotated
code samples and debugging guides to the Zephyr documentation repository
Establish stronger ties between academia and the Zephyr developer community by
encouraging student / instructor participation in forums and open-source contributions
as part of coursework
Establish feedback loops with students and industry partners to ensure the curriculum
remains aligned with evolving best practices and technological advancements</p>
      </sec>
    </sec>
    <sec id="sec-4">
      <title>4. Conclusion and outlook</title>
      <p>
        Integrating Zephyr RTOS into the curriculum and research activities at TU Dresden has
significantly modernized embedded systems education. Its modular design and community support
facilitate quick adaptation for diverse projects and foster practical skill development. Notably, even
students with limited prior experience in embedded systems have successfully engaged with Zephyr,
overcoming initial challenges to design and implement projects at bachelor’s and master’s levels.
This demonstrates Zephyr’s effectiveness in bridging the gap between theoretical knowledge and
real-world application. This demonstrates with the right guidance and resources the challenges of
adopting modern RTOS are surmountable. Nevertheless, our experience has highlighted several
opportunities for further improvement within our courses and labs. The initial learning curve
remains a significant challenge particularly for newcomers to embedded systems. To address this,
we plan to develop more structured onboarding materials e.g. beginner-friendly tutorials in
combination with step-by-step lab guides. Interactive workshops that clarify key concepts like
device trees, KConfig and Zephyr’s build system could also improve the learning experience.
Expanding the peer-to-peer learning and mentoring opportunities can also help students navigate
early hurdles more effectively. Moreover, keeping instructional materials and documentation up to
date is essential in a fast-moving ecosystem. A good example for managing different Zephyr and in
this case SDK versions is the Nordic Semiconductors® developer academy [
        <xref ref-type="bibr" rid="ref9">9</xref>
        ]. Establishing a robust
feedback loop with students and regularly updating course content ensures that our teaching remains
relevant and impactful. The inclusion of real-world case studies or projects from industry can further
bridge the gap between academic learning and professional practice. By constantly expanding
available hardware platforms and project topics in our laboratories, upcoming students will be able
to fully explore the potential of Zephyr and apply their knowledge to a wider range of applications.
The aim is to establish interdisciplinary collaboration and enhance the learning experience to enable
innovative solutions to complex challenges. Although the introduction of Zephyr RTOS at TU
Dresden has already brought significant benefits for students and research there is still considerable
potential for growth but as always that’s a process.
      </p>
      <p>During the preparation of this manuscript, the authors employed DeepL to assist with sentence
refinement and rephrasing. Following the use of this tool, the authors thoroughly reviewed and
edited all content as required and assume full responsibility for the accuracy and integrity of the final
publication.</p>
    </sec>
  </body>
  <back>
    <ref-list>
      <ref id="ref1">
        <mixed-citation>
          [1]
          <string-name>
            <surname>Rivera-Matos</surname>
          </string-name>
          ,
          <article-title>Zephyr RTOS explored: From system design to academic applications</article-title>
          , Zephyr Project,
          <year>2024</year>
          . URL: https://www.zephyrproject.
          <article-title>org/zephyr-rtos-explored-from-system-designto-academic-applications/</article-title>
          <source>(accessed 29.05</source>
          .
          <year>2025</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref2">
        <mixed-citation>
          [2]
          <string-name>
            <given-names>K.</given-names>
            <surname>Albes</surname>
          </string-name>
          ,
          <article-title>Design und Implementierung eines Zephyr-RTOS-Modells für ARA zur statischen Whole-System-</article-title>
          <string-name>
            <surname>Analyse</surname>
          </string-name>
          , Bachelorarbeit, Leibniz Universität Hannover,
          <year>2021</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref3">
        <mixed-citation>
          [3]
          <string-name>
            <given-names>R.</given-names>
            <surname>Rivera-Matos</surname>
          </string-name>
          ,
          <article-title>Zephyr RTOS explored: From system design to academic applications</article-title>
          , Zephyr Project,
          <year>2024</year>
          . URL: https://www.zephyrproject.
          <article-title>org/zephyr-rtos-explored-from-systemdesign-to-academic-applications/</article-title>
          <source>(accessed 29.05</source>
          .
          <year>2025</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref4">
        <mixed-citation>
          [4]
          <string-name>
            <given-names>M.</given-names>
            <surname>Vedral</surname>
          </string-name>
          ,
          <string-name>
            <surname>Zephyr</surname>
            <given-names>RTOS</given-names>
          </string-name>
          , Bachelorarbeit, České vysoké učení technické v Praze,
          <year>2023</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref5">
        <mixed-citation>
          <article-title>[5] Integration of Zephyr RTOS in motor control systems: Challenges and solutions</article-title>
          , in:
          <source>International Journal of Computer Science and Engineering (IJCSE)</source>
          ,
          <year>2024</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref6">
        <mixed-citation>
          [6]
          <string-name>
            <given-names>R.</given-names>
            <surname>Vágó</surname>
          </string-name>
          ,
          <article-title>Development of a safe architecture for embedded systems using Linux and Zephyr RTOS, Master's thesis</article-title>
          , KTH Royal Institute of Technology, Stockholm,
          <year>2022</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref7">
        <mixed-citation>
          [7]
          <string-name>
            <given-names>S.</given-names>
            <surname>Berg</surname>
          </string-name>
          ,
          <string-name>
            <given-names>A.</given-names>
            <surname>Nyffenegger</surname>
          </string-name>
          ,
          <article-title>Connecting constrained devices to the cloud using Zephyr, Bachelorarbeit</article-title>
          , Universität Freiburg,
          <year>2020</year>
          .
        </mixed-citation>
      </ref>
      <ref id="ref8">
        <mixed-citation>
          [8]
          <string-name>
            <given-names>TU</given-names>
            <surname>Dresden</surname>
          </string-name>
          ,
          <article-title>Lehrstuhl für Verkehrstelematik (ITVS)</article-title>
          ,
          <source>Lehrveranstaltungen und Modulbeschreibungen</source>
          ,
          <year>2025</year>
          . URL: https://tu-dresden.de/bu/verkehr/itvs (accessed
          <volume>29</volume>
          .05.
          <year>2025</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref9">
        <mixed-citation>
          [9]
          <string-name>
            <given-names>Nordic</given-names>
            <surname>Semiconductor</surname>
          </string-name>
          ,
          <string-name>
            <surname>Nordic Developer</surname>
          </string-name>
          Academy - Online learning platform,
          <year>2024</year>
          . URL: https://academy.nordicsemi.
          <source>com (accessed 29.05</source>
          .
          <year>2025</year>
          ).
        </mixed-citation>
      </ref>
      <ref id="ref10">
        <mixed-citation>
          [10]
          <string-name>
            <given-names>M.</given-names>
            <surname>Böhme</surname>
          </string-name>
          ,
          <article-title>Empirische Evaluation von Bluetooth LE basierten Abstandsmessungen, Studienarbeit, Technische Universität Dresden, Fakultät für Verkehrswissenschaften "Friedrich List"</article-title>
          ,
          <source>Professur Informationstechnik für Verkehrssysteme (ITVS)</source>
          ,
          <year>2025</year>
          .
        </mixed-citation>
      </ref>
    </ref-list>
  </back>
</article>