Energy-Efficient Sleep Wake-Up Mechanism Based Routing Protocol Using Siamese Network and Optimized Fuzzy Interference System in Green IoT

Abstract

The proliferation of multiple battery-powered sensors under the Internet of Things (IoT) requires the development of sustainable and energy-conscious protocols. However, various energy-efficient mechanisms are developed, but those mechanisms are not much accurate and they have energy loss, transmission collision, and threats to security due to increased data redundancy. To address these issues, the energy-efficient fuzzy-based sleep wake-up mechanism (EEFSWM) was created. Sensor nodes are randomly initialized at the beginning, and the clustering algorithm used is the evidential Gaussian mixture model (EGMM). The cluster head is selected by taking into account each node's energy and distance. To choose the ideal energy and distance, golden jackal optimization (GJO) is employed. The Siamese network is then used to determine whether or not neighbor nodes are perceiving the same data. It computes a similarity measure if it finds related data. The node moves into the sleep stage, while the other nodes go into the waking stage when the similarity measure value surpasses the threshold value. After that, the sleep cycle is computed with a few parameters using the fuzzy inference system (FIS). The node awakens after the update cycle has finished, then joins the clustering and cluster head selection process. This proposed approach attains 7.99-J average residual energy, 84% packet delivery ratio, 2.64% throughput value, and 1105-s network lifetime.