Modelado matemático para la selección de bandas espectrales en la etapa de toma de decisiones en redes de radio cognitiva inalámbricas para un comportamiento de llegada de sus, tipo exponencial.

Abstract

The objective of this research project is to propose and solve an optimization problem that allows the selection and equitable allocation of spectral bands in the decision making stage in a centralized CR network topology. It has as input variables a rate of arrival of SUs exponential type, and QoS requirements, depending on whether the traffic is interactive or massive. The base methodology is the current state of art on the subject of decision making in Cognitive Radio (CR), with this there is a foundation from which the mathematical modeling of prediction of the arrival of the SU to the base station starts. To do this, different proactive techniques that took into account the evaluation of quality of service criteria were analyzed. The method that best fits is data mining, which tries to discover the patterns of large volumes of data. To achieve this, it helps the prediction process evaluate the walls. In addition, it generates a solid system based on the K-means method, the mathematical process is developed and the result of the SU prediction model is obtained. For the channel selection in an equitable way, a dynamic multivariable algorithm for the selection of channels in a CR Centralized network is presented. This channel selection is based on the Fuzzy Analytical Hierarchical process (FAHP). This model has been widely used in decision making where it is necessary to involve a high number of factors and variables, both qualitative and quantitative. A hierarchical model is made, as indicated by the process, and the selected criteria are defined to choose the best channel Real Time (RT) and Best Effor (BE). The evaluation of the subcriteria was developed with the Chang method, the alternatives are selected in view of the objective.