An Energy Aware Routing for Cognitive Radio Wireless Sensor Network using FuzzyQ

Currently, the wireless communication technology has noticed explosive expansion in its demand and it is increasing rapidly whereasdue to their wide range of applications and uses, Wireless Sensor Networks (WSNs) have attracted the attentions of the academic community in various offline and online applications. These networks are deployed in the unattended and harsh environment due to which maintaining the network lifetime becomes a challenging task. The energy consumption in these networks is a challenging task which needs to be addressed for increased network lifetime. Moreover, the conventional wireless communication standards are utilizing the radio spectrum. Currently, the spectrum is very limited and excessive exploitation creating overburdening on the network. Thus efficient spectrum allocation is also becoming a tedious task. Nowadays, Cognitive Radio (CR) has appeared to be a viable way out to efficiently utilize the available spectrum. According to this CR concept, cognitive radio technology helps to opportunistically and dynamically access the spectrum bands which improve the utilization of radio spectrum resources.Several techniques have been presented to deal with the issue but these techniques suffer from various issues such as uncertain energy harvesting and resource (channel) allocation which becomes a critical issue for these networks. To overcome these problems, the workproposes a novel approachcalled ?Fuzzy?_Q-EACWSNR (an energy aware cognitive wireless sensor network routing) scheme which minimizes the energy consumption by using reinforcement based Q-learning scheme for packet forwarding in cognitive enabled wireless sensor nodes.This novel scheme considers the optimal cluster formulation, cluster head selection, spectrum sensing and allocation.The experimental study reported thenoteworthy improvement in the communication performance by evaluating and comparing the performance with EACRP, ESAC andOptimum distance based clustering in terms of network throughput, average packet delay, and energy consumption.