Modelo de handoff espectral proactivo-predictivo que permite reducir el nivel de interferencia en redes de radio cognitiva

Abstract

Cognitive radio is a technique that is revolutionizing the world of wireless connections because it proposes, through a Dynamic Access Model to the Spectrum or DSA, that Users who are on saturated frequencies can access other channels opportunistically without affecting the owner user. Being important to avoid interference, this document addresses a proactive-predictive handoff model for the transmission of unlicensed users and diminish the collision with licensed users, through multivariable techniques such as SAW, TOPSIS, VIKOR and AHP for the selection of channel and Time Series MA, AR, ARMA for the prediction of arrival of the licensed user; the model is compared with reactive Handoff model. The analysis is made on High and Low traffic for traces captured in the city of Bogotá, Colombia and it is shown that depending on the traffic the interference decreases between 5 and 15%.