Cognitive radio networks management using an ANFIS approach with QoS/QoE mapping scheme

Future networks are characterized by a panoply of novel services based on multimedia services like gaming and real time video streaming. In addition, cognitive radio is considered as an emergent candidate of next generation (NextG) networks. Therefore, there is a need of some techniques that can ensure self-managed networks with learning capabilities. Our approach is based on adaptive neuro-fuzzy inference system (ANFIS) used for predicting the user video perception (e.g. MOS) and for managed decisions that can be achieved by a specific radio configuration (e.g. data rate, handover). The ANFIS model with Quality of Services/Quality of Experience (QoS/QoE) mapping is able to sense environment, decide, learn and optimize its decisions online by a learning algorithm that uses a set of experimental measurements. We used an implementation tool of the ANFIS model under MATLAB/SIMULINK environment supporting the development of real time scenarios.