BMREWSN: Design of a hybrid Bio-inspired Model for improving the Routing performance of Energy-aware Wireless Sensor Networks

Designing low-power wireless sensor networks is a multidomain task that involves estimation of node and network parameters, identification of routing redundancies, optimizing sleep scheduling cycles, improving data processing capabilities, and grouping nodes into performance-aware clusters. To perform these tasks, researchers have developed various models, most of which work on optimization of a limited set of parameters. Models that aim at optimizing multiple parameters are observed to have higher computational complexity, which reduces their energy efficiency under real-time use cases. Due to these lacunas, their applicability and scalability performance are reduced when applied to large-scale network scenarios. To overcome these limitations, this paper discusses the design of a novel hybrid bio-inspired model for improving the routing performance of energy-aware wireless sensor networks. The model initially deploys a Genetic Algorithm (GA) based sleep scheduling technique, which is able to optimize the energy levels of network nodes via intelligent duty cycle control operations. This is cascaded with a Grey Wolf Optimization (GWO) Model, which assists in delay-aware, throughput-aware, and energy-aware routing under large scale network scenarios. The model uses an intelligent node selection technique due to which the number of computations needed for evaluation of GA and GWO is reduced exponentially. This reduction in computation allows the model to work with a reduced set of nodes, thus increasing the lifetime for multiple network types. The QoS (Quality of Service) was measured and compared with various state-of-the-art methods. Based on this comparison, it was observed that the proposed The model is capable of reducing communication delays by 14.5%, improving throughput by 8.3%, reducing energy consumption by 8.5%, and improving packet delivery ratio by 3.2% under different communication scenarios. Due to these enhancements, the proposed model is capable of being deployed for a wide variety of large-scale network use cases.