**6. Conclusions**

We have presented the design and implementation of a waste managemen<sup>t</sup> system for Smart Campuses. The system is based on a smart waste bin prototype, an innovative device that can be used for automatically recycling objects using a hybrid sensor/image-based classifier. Results showed that the proposed approach reached an accuracy of over 97% for waste classification. In addition, we evaluated the device in different network scenarios, including moving the artificial intelligence on the MEC of the 5G network, reducing the recognition latency and the energy consumption. Moreover, we presented an application server which is able to easily monitor the status of waste bins present in the campus, as well as optimizing the managemen<sup>t</sup> procedures (e.g., waste collection). We believe such a system will be extremely useful in the near future, considering the increased environmental impact of waste generated by people, which requires correct recycling. For this reason, we plan to create many other smart waste bin devices and deploy them on the university premises. This will also enable the possibility to study the interaction between students and smart waste bins, paving the way for possible future system optimizations.

**Author Contributions:** Conceptualization, A.E.C.R., M.B. and P.B.; methodology, A.E.C.R., M.B. and P.B., software, E.L., F.A.S.; validation, E.L. and F.A.S.; writing—original draft preparation, E.L. and A.E.C.R.; writing—review and editing, P.B.; supervision, A.E.C.R. and M.B.; project administration, S.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research work has been supported by project BASE5G (project id: 1155850) funded by Regione Lombardia within the framework POR FESR 2014-2020 and by Vodafone Italia S.p.A under the Italian 5G trial use case *Smart City and Smart Campus*.

**Data Availability Statement:** The waste dataset and the Smart Waste Bin simulator used in this work are available at https://tinyurl.com/SWB-dataset and https://tinyurl.com/SWB-sim, (accessed on 22 May 2021).

**Acknowledgments:** The authors would like to thank Angelo De Iesi, for their invaluable help in the product design of the Smart Waste Bin, and Sabrina Baggioni, Gianpiero Carocci, Giacoma Caruso, Andrea Ferrara and Stefano Bauro from Vodafone Italia S.p.A. for their fruitful ideas.

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