Leonardo Muñoz
Leonardo Muñoz
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Leonardo Muñoz has a total of 7 co-authors. Here are the top co-authors by number of shared publications:
- Francisca V. Vera Vera
Department of Electrical Engineering, Universidad de Concepción, Edmundo Larenas 219, Concepción 4030000, Chile
1 shared publication - Francisco Perez
Electrical Engineering Department, University of Concepci´on, Concepci´on, Chile
1 shared publication - Gabriel Saavedra
Department of Electrical Engineering, Universidad de Concepción, Edmundo Larenas 219, Concepción 4030000, Chile
1 shared publication - Lien Rodríguez-López
Facultad de Ingeniería, Universidad San Sebastián, Lientur 1457, Concepción 4030000, Chile
1 shared publication - Lisandra Bravo Alvarez
Department of Electrical Engineering, Universidad de Concepción, Edmundo Larenas 219, Concepción 4030000, Chile
1 shared publication - Samuel Montejo-Sánchez
Instituto Universitario de Investigacin y Desarrollo Tecnolgico, Universidad Tecnolgica Metropolitana (UTEM), Santiago, Chile
1 shared publication
Latest Publications
- Open Access
High-Accuracy Deep Learning-Based Detection and Classification Model in Color-Shift Keying Optical Camera Communication Systems
The growing number of connected devices has strained traditional radio frequency wireless networks, driving interest in alternative technologies such as optical wireless communications (OWC). Among OWC solutions, optical camera communication (OCC) stands out as a cost-effective option because it leverages existing devices equipped with cameras, such as smartphones and security systems, without requiring specialized hardware. This paper proposes a novel deep learning-based detection and classification model designed to optimize the receiver’s performance in an OCC system utilizing color-shift keying (CSK) modulation. The receiver was experimentally validated using an LED matrix transmitter and a CMOS camera receiver, achieving reliable communication over distances ranging from 30 cm to 3 m under varying ambient conditions. The system employed CSK modulation to encode data into eight distinct color-based symbols transmitted at fixed frequencies. Captured image sequences of these transmissions were processed through a YOLOv8-based detection and classification framework, which achieved % accuracy in symbol recognition. This high precision minimizes transmission errors, validating the robustness of the approach in real-world environments. The results highlight OCC’s potential for low-cost applications, where high-speed data transfer and long-range are unnecessary, such as Internet of Things connectivity and vehicle-to-vehicle communication. Future work will explore adaptive modulation and coding schemes as well as the integration of more advanced deep learning architectures to improve data rates and system scalability.Open AccessHigh-Accuracy Deep Learning-Based Detection and Classification Model in Color-Shift Keying Optical Camera Communication Systems
The growing number of connected devices has strained traditional radio frequency wireless networks, driving interest in alternative technologies such as optical wireless communications (OWC). Among OWC solutions, optical camera communication (OCC) stands out as a cost-effective option because it leverages existing devices equipped with cameras, such as smartphones and security systems, without requiring specialized hardware. This paper proposes a novel deep learning-based detection and classification model designed to optimize the receiver’s performance in an OCC system utilizing color-shift keying (CSK) modulation. The receiver was experimentally validated using an LED matrix transmitter and a CMOS camera receiver, achieving reliable communication over distances ranging from 30 cm to 3 m under varying ambient conditions. The system employed CSK modulation to encode data into eight distinct color-based symbols transmitted at fixed frequencies. Captured image sequences of these transmissions were processed through a YOLOv8-based detection and classification framework, which achieved % accuracy in symbol recognition. This high precision minimizes transmission errors, validating the robustness of the approach in real-world environments. The results highlight OCC’s potential for low-cost applications, where high-speed data transfer and long-range are unnecessary, such as Internet of Things connectivity and vehicle-to-vehicle communication. Future work will explore adaptive modulation and coding schemes as well as the integration of more advanced deep learning architectures to improve data rates and system scalability.