Compatibilidad de red híbrida VLC-RF femtoceldas con Python mediante algoritmos para la mejora de recursos.
Loading...
Date
2025
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Universidad de Concepción
Abstract
Ante la creciente saturación del espectro de radiofrecuencia (RF), las redes híbridas VLC-RF emergen como una solución que requiere una gestión de recursos inteligente. El objetivo de este trabajo fue implementar y evaluar tres algoritmos de selección de red (secuencial, analítico y diferencial) para optimizar el rendimiento de un sistema híbrido simulado. La metodología consistió en implementar los algoritmos en el simulador de eventos discretos VlcSim y evaluarlos en cuatro escenarios que variaban la densidad de usuarios y la demanda de capacidad. El rendimiento se analizó mediante métricas clave como throughput, latencia y distribución de la carga.
Los resultados demostraron que no existe un algoritmo universalmente óptimo, pues su rendimiento depende del escenario de uso. El algoritmo analítico se mostró como el más robusto y equilibrado, especialmente bajo alta carga. En contraste, el secuencial resultó inestable bajo congestión, y los diferenciales evidenciaron un fuerte trade-off, maximizando el throughput a costa de la calidad de servicio. Se concluye que la elección del algoritmo es una decisión de diseño estratégica, que debe adaptarse a los objetivos específicos de la red, ya sea priorizar la estabilidad o la velocidad de transferencia total.
In response to the growing saturation of the radio frequency (RF) spectrum, hybrid VLC-RF networks emerge as a solution requiring intelligent resource management. The objective of this work was to implement and evaluate three network selection algorithms (sequential, analytical, and differential) to optimize the performance of a simulated hybrid system. The methodology involved implementing the algorithms in the VlcSim discrete-event simulator and testing them across four scenarios with varying user density and capacity demands. Performance was analyzed using key metrics such as throughput, latency, and load distribution. The results showed that no single algorithm is universally optimal, as its performance depends on the use-case scenario. The analytical algorithm proved to be the most robust and balanced, especially under high load. In contrast, the sequential algorithm became unstable under congestion , and the differential algorithms showcased a stark trade-off, maximizing throughput at the expense of quality of service. It is concluded that the choice of algorithm is a strategic design decision that must be adapted to the network’s specific objectives, whether prioritizing stability or total transfer speed.
In response to the growing saturation of the radio frequency (RF) spectrum, hybrid VLC-RF networks emerge as a solution requiring intelligent resource management. The objective of this work was to implement and evaluate three network selection algorithms (sequential, analytical, and differential) to optimize the performance of a simulated hybrid system. The methodology involved implementing the algorithms in the VlcSim discrete-event simulator and testing them across four scenarios with varying user density and capacity demands. Performance was analyzed using key metrics such as throughput, latency, and load distribution. The results showed that no single algorithm is universally optimal, as its performance depends on the use-case scenario. The analytical algorithm proved to be the most robust and balanced, especially under high load. In contrast, the sequential algorithm became unstable under congestion , and the differential algorithms showcased a stark trade-off, maximizing throughput at the expense of quality of service. It is concluded that the choice of algorithm is a strategic design decision that must be adapted to the network’s specific objectives, whether prioritizing stability or total transfer speed.
Description
Tesis presentada para optar al título de Ingeniero/a Civil en Telecomunicaciones.
Keywords
Sistemas de transmisión de datos, Diseño de sistemas, Algoritmos computacionales