Algoritmos para Toma de Decisiones en Redes Inalámbricas Cognitivas: una Revisión

Abstract

El objetivo de este artículo es presentar una revisión de técnicas basadas en inteligencia artificial para el proceso de toma de decisiones en redes de radio cognitiva. Las redes de radio cognitiva surgieron como una solución para resolver los problemas de asignación fija y de escasez del espectro, trabajan con un modelo de gestión que se denomina ciclo cognitivo, donde la toma de decisiones es clave, ya que permite seleccionar la oportunidad espectral más adecuada. A partir de literatura actualizada, se analizan estrategias que implementan máquinas de aprendizaje automático, algoritmos bioinspirados, teoría de juegos y algoritmos de consensos. Finalmente, el documento presenta los retos para el proceso de toma decisiones, donde se resalta la necesidad de seguir aprovechando los avances en inteligencia artificial para obtener mejores resultados. Sin embargo, las estrategias deben ser escalables para facilitar la carga computacional y poder resolver problemas de mayor complejidad.

Spectral decision for cognitive radio networks in a multi-user environment

“…CR emerged 20 years ago as a technology to overcome scarcity problems using dynamic access to the spectrum. It is characterized by perceiving, learning, planning (decision making) and acting according to current network conditions (Giral et al, 2019;Wang et al, 2020a). In cognitive radio networks (CRNs), unlicensed users (SU) must make smart decisions based on spectrum variation and actions taken by other SU.…”

“…However, there is no single best technique; each structure has a considerable number of advantages and disadvantages, which makes it possible to characterize them for specific decision-making problems. A classification of decision-making algorithms in CRN is described in (Giral et al, 2019;Rizk et al, 2018;Wang et al, 2020a).…”

Evaluation of the Failed Handoff Rate in Collaborative Wireless Networks

“…Spectrum decision-making is a key function of CRN. However, it has not been as researched in comparison to other functions of the cognitive cycle [7,[12][13][14]. The decision-making process aims to choose the best spectral opportunity among a finite set of possibilities, allowing the SU to generate a sequence of actions that will lead to the completion of its objectives [13][14][15][16].…”

“…However, it has not been as researched in comparison to other functions of the cognitive cycle [7,[12][13][14]. The decision-making process aims to choose the best spectral opportunity among a finite set of possibilities, allowing the SU to generate a sequence of actions that will lead to the completion of its objectives [13][14][15][16]. An inadequate decisionmaking process can significantly affect the parameters of quality, latency, throughput, reliability, signalization, PU interference, energy efficiency, bandwidth, SINR and error rate [2,[17][18][19][20].…”

“…Collaborative strategies have delivered new models for the efficient use of radio resources, as well as the decision-making process for CRN. In collaborative decision-making, users can communicate between them to exchange local measurements regarding availability and interference, as well as other variables [14]. The purpose is to take advantage of spatial diversity which is achieved by the collaborative of unlicensed users who share information with neighbors [21].…”