A robust conscious model for enhancing cognitive radio quality of service
Ayodele Abiola Periola, Olabisi Emmanuel Falowo · 2016
Cognitive radios (CRs) use learning algorithms (LAs) located in the cognition engine (CE) to adapt their behaviour. CRs use LAs for spectrum prediction to enhance their quality of service (QoS). The CR's CE consumes battery power while classifying LAs. The LA classification reduces CR data transmission power and limits CR throughput. This paper proposes a framework to enhance CR QoS in LTE-Advanced (LTE-A). The proposed framework reduces battery power expended in LA classification and increases CR data transmission power. It introduces the radio resource control (RRC); RRC_COGNITIVE state in which the CR pauses LA classification. The framework's performance is evaluated using the CR transmit power and throughput. Simulations show that the proposed framework reduces LA classification power by 65 % on average. The reduction of LA classification power enhances CR transmit power. The CR throughput is enhanced by 23% and 80% when CRs are and are not secondary users (SUs) respectively.