**5. Conclusions**

In this paper, we presented a people-number estimation algorithm based on nonintrusive, sparse-distributed sensors data from a multi-resident smart environment with work on the continuous flow of data generated by the sensors. Estimating the exact number of people in a family with more than two residents is a difficult task, especially in a sparsedistributed sensor network where each room has only one binary sensor to detect the presence of a human. However, the choice for such a setting, which is basic and minimal, is affordable in practice in many situations. Moreover, having a good, even if not precise, estimation of the number of people can be sufficient in many real scenarios of older people living alone at home.

Our algorithm has several advantages: it does not need to learn any data and therefore can be applied immediately, starting at any time, and only limited information about the house settings is needed. A good accuracy has been obtained thanks to the representation of the status of the rooms and the multi-branch inference based on the context.

As future work, we plan to also test other types of sensor data, such as bed/chair sensors, to evaluate the results in motionless situations [55] where no PIR sensors are activated.

**Author Contributions:** Conceptualization, F.S.; Methodology, A.M.; Supervision, S.C. and F.S.; Software, C.L.; Validation, C.L. and A.M.; Writing—original draft, C.L.; resources, F.S.; data curation, F.S. and C.L.; Writing—review and editing, S.C., F.S and A.M. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research received no external funding.

**Institutional Review Board Statement:** Not applicable.

**Informed Consent Statement:** Not applicable.

**Data Availability Statement:** The data presented in this study are available on request from the corresponding author. The data are not publicly available due to privacy reasons.

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