Naslov
Context-based System for User-Centric Smart Environment

Autori
Mandaric, Katarina ; Skocir, Pavle ; Jezic, Gordan

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
2020 International Conference on Software, Telecommunications and Computer Networks (SoftCOM) / - Split : Institute of Electrical and Electronics Engineers (IEEE), 2020

ISBN
978-953-290-099-6

Skup
28th International Conference on Software, Telecommunications and Computer Networks (SoftCOM 2020)

Mjesto i datum
Hvar, Hrvatska; online, 17.09.2020. - 19.09.2020

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
Cognitive Internet of Things, Artificial Neural Networks, User-Centricity, Smart Lighting

Sažetak
Internet of Things (IoT) solutions are becoming ubiquitous in various application domains. In homes and offices, IoT systems enable remote and scheduled operation of heating, ventilation and air conditioning (HVAC) systems, lighting control, control of kitchen and other home appliances, etc. Available solutions in the smart home domain mainly enable controlling systems and appliances remotely, e.g., by using a smartphone, or manually setting the times in which systems and appliances should operate. In this paper we propose a system for indoor lighting control which sets the colors and brightness of the lamps based on the context: time of the day, current lighting conditions in the room, and having users in the focus - their lighting preferences in the room. Feed-forward neural networks are used for learning about the most pleasing settings. The major advantage of this system is its user- centricity, i.e., taking into account users' color and brightness preferences. The system is evaluated on twenty-two (22) users in a laboratory environment and demonstrates general user score rate of 68.18% as "great" and 31.82% as "good".

Izvorni jezik
Engleski

Znanstvena područja
Elektrotehnika, Računarstvo

POVEZANOST RADA