EE-FNN: An Intelligent Hybrid and Fuzzy Neural Network Framework for Energy-Efficient Cluster Head Selection to Enhance Learning Efficiency in Special Education

Abstract

Wireless Sensor Networks (WSNs) are essential for many different applications, such as smart cities, healthcare, industrial automation, surveillance for the military, and environmental monitoring. Motivated by microelectronics progress, WSNs use low-power circuits to maximize energy consumption—an important consideration considering the short battery life of sensor nodes. In order to improve energy efficiency in WSNs, this research looks at several optimization techniques. It compares the EE-FNN protocol against OEERP, LEACH, BCDCP, and DRINA. The results show that, with an initial energy consumption of 7.8 J and a maximum energy consumption of 46.8 J, EE-FNN continuously exhibits the lowest energy consumption, demonstrating its better energy management skills. Furthermore, EE-FNN achieves a Packet Delivery Ratio (PDR) of 99.7% to 100% and maintains throughput rates between 77,605 bps and 76,335 bps, demonstrating exceptional data transmission efficiency. On the other hand, protocols such as DRINA and BCDCP show worse data transmission performance and increased energy usage. The research also emphasizes how well EE-FNN can maintain network lifetime, as demonstrated by the fact that it can operate from 11,086 rounds to 1,847 rounds, much surpassing the performance of other protocols. These findings demonstrate how important energy-efficient protocols like EE-FNN are to improving WSN performance, ensuring trustworthy data transfer, and extending the lifespan of networks.