Bio-Inspired Based Optimization of Machine Learning Techniques for Energy Efficient Routing for UWSN

Underwater Wireless Sensor Networks (UWSN) is one of the most emerging branches of wireless sensor networks. They are used extensively in environmental monitoring, oceanography, and marine biology. Energy efficiency, signal attenuation, and changeable climatic conditions are just a few of the major obstacles these networks must overcome. The criteria for their functionality differs a lot from a basic wireless sensor network. UWSN are way more complicated when compared to basic wireless sensor networks. Energy Efficient Multipath Routing allows the functioning of UWSN in a highly efficient manner with higher energy efficiency and delivery ratio. This paper proposes a hybrid energy-efficient routing protocol for UWSNs that combines neural network approaches with bio-inspired algorithms. firefly optimization, kookaburra optimization, hippopotamus optimization algorithms from bio-inspired based are used to achieve energy efficiency. Comprehensive performance analysis show that the firefly optimization based hybrid achieves better accuracy and energy efficiency.