Cognitive decision engine design for CR based IoTs using Differential Evolution and Bat Algorithm

Current research efforts in communication technology are shifting towards new paradigm namely, Internet of Things (IoTs). There is a strong need to tackle the challenge of introduction of massive data into network by IoT supported applications. For this, Cognitive Radio networks (CRNs) are seen as a potential solution. Enabling IoT objects with Cognitive radio features has led to new research dimension of CR based IoTs. Real time tuning of transmission parameters by Cognitive decision engine as per user needs and dynamic environment conditions is one of the important tasks. Determination of optimal value of transmission parameters becomes even more challenging for a multicarrier system because of high dimensionality as there are large number of decision variables to be optimized. Nature inspired metaheuristic optimization techniques offer an efficient and simple solution to the aforementioned problem. In this paper, comparative performance analysis of Differential evolution (DE) and Bat algorithm has been done for the parameter tuning problem. The results demonstrate that the parameter adaptation by DE based engine outperforms the Bat based implementation in terms of fitness score for the five different transmission modes supported by CR based IoTs.