Enhanced Data Security in Wireless Sensor Networks Through Multi-Objective Energy-Aware Routing Utilizing Interpretable Generalized Additive Neural Network Approach

In Wireless Sensor Networks (WSNs), multiple-objective cluster head (CH) depending on energy-aware routing seeks to maximize data aggregation while ensuring security and energy efficiency. To balance energy consumption along with network durability, this strategy entails choosing CHs based on a count of factors, such as energy levels, node distance, and data security. Longer processing times and more computing overhead might result from the difficulty of maximizing numerous objectives. Furthermore, predefined models and heuristics may not always work well in dynamic WSNs due to factors like variable node densities and energy levels. To overcome these issues, in this paper Enhanced Data Security in WSNs through Multi-Objective Energy-Aware Routing Utilizing Interpretable Generalized Additive Neural Network Approach (OEWSN-MA-IGANN) is proposed. Thus, data aggregation basis Interpretable Generalized Additive Neural Network (IGANN) is employed to choose CH utilizing a multiple-objective fitness function. Energy, latency, output, distance between nodes, capacity, collision, traffic rate, and cluster density are among the variables taken into consideration by this fitness function. Then, a Bitterling Fish Optimization (BFO) is proposed to optimize the weight parameters of IGANN. The proposed OEWSN-MA-IGANN approach is implemented in Python and evaluated with some metrics, like packet delivery ratio (PDR), network lifetime, computational time, delay, scalability, network load, network size, and overhead. The simulation outcomes depict that the proposed OEWSN-MA-IGANN approach reaches 23.12%, 24.14%, and 21.08% higher network lifetime, 21.12%, 22.11%, and 21.06% higher PDR are analyzed with existing methods.