AUGMENTED Q-LEARNING-BASED ARTIFICIAL BEE COLONY ALGORITHMS IN WIRELESS SENSOR NETWORKS LEVERAGING DEEP LEARNING FOR ENERGY OPTIMIZATION

Abstract

Routing with multiple hops is employed by Wireless Sensor Networks (WSNs) to facilitate efficient data transfer; yet, Use of energy is still a significant issue for maintaining dependable connectivity. Minimizing energy consumption and enhancing network interactions are crucial for the sustained viability of wireless sensor networks (WSNs). Limitations on energy impede the efficacy of data's sensor nodes (SNs) Transmission throughout the network, not withstanding the benefits of multiple-hop routing. The difficulty is in identifying the best strategy to reduce the amount of energy utilize while preserving dependable data transfer. We proposed an ideal approach when choosing a Cluster Head (CH) in Wireless Sensor Networks (WSNs)utilizing an enhanced Algorithm for Artificial Bee Colonies (IQ-ABC) based on Q-learning to enhance multi-hop routing's effectiveness in WSNs. This study introduces an augmented ABC algorithm that incorporates Q-learning to improve the stages of Investigating and exploiting. The exploitative capacity of the IQ-ABC is enhanced by an altered Q-learning process. The IQ-ABC approach recognises the optimal energy-efficient route aimed at each sensor node to communicate data to the collection inside the suggested system. Additionally, a multi-objective fitness function enhances the selection of cluster heads through weighted assignment utilizing fuzzy logic, balancing critical factors like as trust, latency as well as energy efficiency. The simulation findings indicate that, relative to traditional routing algorithms, the IQ- ABC technique greatly lowers energy usage besides helps to enhance the longevity of nodes for sensors. The paper concludes by emphasizing the necessity of optimizing computer resources to sustain how well machine learning works techniques in detecting abnormalities in the network. This paper examines the innovative incorporation of deep learning applications in the deployment of Networks of Wireless Sensors(WSN).