FIND-ROUTE: Fourier series integrated deep learning model for energy efficient routing in Internet of Things-wireless sensor network

Abstract

The Internet of Things (IoT) relies on wireless sensor networks (WSNs) to transmit data across a wide range of applications. However, the commonly encountered primary challenges in IoT-enabled WSNs are high energy consumption during data transmission, which insists energy optimized routing to prolong the network lifetime. To address these challenges, a novel Fourier series integrated deep learning-based routing (FIND-ROUTE) framework has been proposed for energy-aware communication among IoT nodes in WSN. Initially, a hybrid clustering approach forms an adaptive cluster for efficient data aggregation with reduced energy consumption. After clustering, stable cluster heads (CHs) are elected by a Fourier series-based metaheuristic optimization algorithm for balancing the energy usage with extended network lifetime. Finally, an Intelligent neural network dynamically selects the optimal path and transmits the data efficiently with reduced latency for reliable communication in IoT-WSN. The FIND-ROUTE framework is simulated by using MATLAB, and it is validated by using the WSN-DS dataset. The proposed FIND-ROUTE framework is evaluated based on several parameters, including energy consumption, packet delivery ratio (PDR), network lifetime (NL), time complexity, throughput, number of alive nodes, packet loss ratio (PLR), and space complexity. In comparison, the proposed FIND-ROUTE framework achieves a PDR of 90%, whereas MLBDARP, LQEER, and NBSHO-DRNN achieve 70%, 60%, and 67% respectively.