Identificador persistente para citar o vincular este elemento: http://hdl.handle.net/10553/128891
Título: Color space-based autoencoder for optical camera communications
Autores/as: Luna-Rivera, J. M.
Rabadan, J. 
Rufo, J. 
Guerra, V. 
Moreno, D.
Perez-Jimenez, R. 
Clasificación UNESCO: 3325 Tecnología de las telecomunicaciones
Palabras clave: Artificial Intelligence
Autoencoder
Csk
Occ
Vlc
Fecha de publicación: 2024
Publicación seriada: Expert Systems with Applications 
Resumen: This paper proposes an end-to-end optical camera communications (OCC) system using an autoencoder neural network trained to recover the transmitted symbols. Although OCC techniques have been extensively studied in the literature, using an autoencoder that learns the transmitter and receiver functions jointly is a novel concept with significant prospects. Furthermore, we investigate the performance impact caused by the overlooked optical-to-electrical (O2E) conversion process of real-world OCC receivers. The autoencoder learning model captures these typically undesired changes in image sensors for the design of constellation symbols and reception schemes. For the simulation, we constructed an end-to-end autoencoder for a color space-based OCC system and measured the O2E performance effect. The proposed autoencoder communication system is analyzed and compared using the symbol error rate (SER) across various OCC detection systems. Despite the subtle spectral responsivity variations in image sensors, our numerical results indicate that the autoencoder model can learn to recover the transmitted data while minimizing SER and meeting the lighting requirements. These findings may interest a broad range of applications, particularly in IoT sensor networks. Among all the image sensors we studied, the OCC system with Bayer CFA-based signal detection showed superior performance.
URI: http://hdl.handle.net/10553/128891
ISSN: 0957-4174
DOI: 10.1016/j.eswa.2023.123101
Fuente: Expert Systems with Applications[ISSN 0957-4174],v. 245, (Julio 2024)
Colección:Artículos
Vista completa

Visitas

61
actualizado el 31-ago-2024

Google ScholarTM

Verifica

Altmetric


Comparte



Exporta metadatos



Los elementos en ULPGC accedaCRIS están protegidos por derechos de autor con todos los derechos reservados, a menos que se indique lo contrario.