**8. Conclusions**

The learning/teaching processes in the development cycle of IoT solutions imply a set of skills ranging from devices and IoT sensors, their communication protocols, the storage managemen<sup>t</sup> and the processing environments on the Cloud for data generated by sensors. These environments are then eventually able to make decisions or show the relevant information on those sensors (as indicators). These fundamental competences are needed in the full cycle of development of IoT solutions, consisting on three layers: (1) basic interaction with sensors and specific communication protocols; (2) data managemen<sup>t</sup> models to handle the generated data; and (3) processing and visualization of the most relevant indicators on these IoT devices. In this last step, the processing can include a specific communication protocol. This protocol could be used to perform actions in the IoT device itself as a response to the processed indicators (for example, using available actuators at the device).

According to this, this work first presents the main features of the LoT@UNED platform, which has been developed to cover the instructional design of our subjects, and how the three layers of the proposed full cycle of development for IoT solutions are implemented in it. The essential characteristics for this kind of laboratories/environments are fulfilled by this platform: edge programming, fog programming, cloud dashboard and analytics programming, protocol experimentation and cybersecurity. Each phase is associated with a specific activity that is deployed in a standard way using Docker containers managed through a cluster manager (with Kubernetes). The manager balances the workload of different devices. Thus, the use of the devices/sensors is assigned in a dynamic way to the students who are developing the activities. This platform allows students to implement all these phases efficiently and redundantly, providing high availability for its use.

The proposed LoT@UNED platform has also been used for students in several computer science subjects. The use of this platform is especially relevant in online educational environments, as is the case of distance universities. This way, they perform remote experimental activities with a collaborative IoT learning infrastructure in the cloud, analyze the data generated and make visual representations in it. As for the result and discussion sections, we can conclude that the perceived usefulness and ease of use of the proposed platform values are really good, as well as the intention of use it in the future for additional practices. The students' attitude is also grea<sup>t</sup> with respect to the use of the platform in practical activities. The rest of the indicators are good, although they are challenging for working on improving the social influence among students when using it, and easing the access mechanisms.

As for future work, the presented method for validation of the IoT platform will be improved. To achieve this, a UTAUT model will be hypothesized. The same set of factors will be considered (easy of use, usefulness, attitude, social influence, . . . ) to be included in this model, in order to check the intention to use the presented technology. Another future line of research is to exhaustively analyze the students' learning progress into the LoT@UNED platform. Finally, the source code of this tool has not ye<sup>t</sup> been shared with any other institution but the release of the code is also one of our next steps for future work. We would like to have a community around it to go on including improvements.

**Author Contributions:** Conceptualization, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Data curation, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Formal analysis, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Funding acquisition, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Investigation, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Methodology, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Project administration, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Supervision, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Validation, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Visualization, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Writing—original draft, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C.; Writing—review & editing, R.P.-V., L.T., A.R.-G., S.M., R.H. and J.C. All authors have read and agreed to the published version of the manuscript.

**Funding:** This work is a part of the eNMoLabs research project, and it is funded by the Universidad Nacional de Educación a Distancia (UNED).

**Acknowledgments:** The authors would like to acknowledge the support of the eNMoLabs research project for the period 2019–2020 from UNED; our teaching innovation group, CyberGID, started at UNED in 2018 and its associated CyberScratch PID project for 2020; another project for the period 2017–2018 from the Computer Science Engineering Faculty (ETSI Informática) in UNED; and the Region of Madrid for the support of E-Madrid-CM Network of Excellence (S2018/TCS-4307). The authors also acknowledge the support of SNOLA, officially recognized Thematic Network of Excellence (RED2018-102725-T) by the Spanish Ministry of Science, Innovation and Universities.

**Conflicts of Interest:** The authors declare no conflict of interest.
