An Attentive LSTM with Fuzzy Network for Intelligent Optimal Routing in Wireless Sensor Networks using Enhanced Red Piranha-based Optimal Path Selection Mechanism

Abstract

Abstract Several agricultural processes can be greatly improved by data mining and sensor networks approaches. This action is controlling the amount of water used in agricultural crops. Furthermore, the Wireless Sensor Networks (WSN) have emerged as a more important technology in forming field. Utilizing energy and extending the lifespan of the sensor nodes is a key component of WSN design. The enormous problem of energy consumption reduction is used to increase the lifetime of the entire framework. Smart sensing IoT devices typically have low batteries and are not usually rechargeable. The effective protocols seek to extend structure life span while reducing the power consumption to prevent disruptions to the nodes installed for environment controlling and data collection. An intelligent deep learning-based routing in WSN is developed to enhance the network lifetime. This model provided efficient optimal routing, in which the routing decisions are taken by Attentive LSTM with Fuzzy (ALSTMFuzzy). Here, the routing decisions are selected based on the security, malicious node, link availability and scalability, etc. The effective routing is performed by the optimal path selection using Mutated Fitness-based Red Piranha Optimization (MFRPO). The path selection for intelligent routing is performed based on several multiple constrains like distance, delay, PDR, energy consumption and throughput. The developed intelligent routing in WSN system outcome is compared to existing models with effectiveness measures.