Te invitamos a la charla presencial impartida por el Dr. Carlos Fernández Hernández, exalumno USM e Investigador Postdoctoral en Inria Lyon / INSA Lyon (Francia).
🗓 ¿Cuándo? Viernes 6 de noviembre | 12:20 a 13:30 hrs.
📍 ¿Dónde? Sala de clases del DIE.
🍕 ¿Bonus? Colación incluida para asistentes inscritos.
🔗 Inscríbete aquí para asegurar tu colación:
https://indico.ing.puc.cl/event/355/registrations/309/
Resumen
LoRaWAN is a widely adopted technology for low-power, long-range Internet of Things (IoT) applications. Despite its bidirectional design, communication in LoRaWAN networks is predominantly uplink-oriented, while downlink transmissions remain constrained by gateway limitations, duty-cycle regulations, and half-duplex operation. These constraints make downlink communication a significant bottleneck, despite its importance for device activation, acknowledgements, and actuator control.
While existing research has extensively addressed uplink communication and gateway-level constraints, less attention has been devoted to the role of the network server in downlink management. The network server makes critical decisions, including gateway selection and transmission timing, that directly influence downlink performance. This work shifts the focus to the network server by evaluating the downlink scheduling mechanisms of two widely used open-source LoRaWAN network servers, ChirpStack and The Things Stack (TTS), both qualitatively and quantitatively. By comparing their performance against an optimal scheduling model, the evaluation reveals significant room for improvement. Building on these findings, four complementary scheduling approaches are investigated, leveraging prioritization criteria, weighted decision mechanisms, traffic prediction, and duty-cycle management. These approaches aim to improve downlink performance while maintaining compatibility with the LoRaWAN standard and limiting computational complexity.
The findings highlight the importance of network-server decisions in improving downlink capacity and network performance. They also motivate further research into adaptive scheduling, resource allocation, and emerging deployment scenarios, including Direct-to-Satellite IoT, where different communication constraints introduce new challenges for downlink management.
Biografía del Expositor
Carlos Fernández Hernández is a postdoctoral researcher at the CITI Lab, jointly affiliated with Inria Lyon and INSA Lyon, France. He holds a degree in Electronic Civil Engineering from Universidad Técnica Federico Santa María, Chile, and a Ph.D. in Computer Science from INSA Lyon. His research focuses on low-power wide-area networks (LPWANs), particularly LoRaWAN, downlink communication, and direct-to-satellite IoT communications.
¡Los esperamos!
--
Assistant Professor
Electrical Engineering Department
UC Chile