Wireless Sensor Networks (WSNs) have employed in recent years for many different applications and functions. But, it has the critical task to detect the malicious node because node malicious attacks are dangerous attacks, and the concept of a malicious attack is opponents enter the network, search accidentally, and capture one or more normal nodes. A lot of research developed to overcome this problem, but no precise results are found. In this paper, design a Hybrid Vulture and African Buffalo with Node Identity Verification (HVAB-NIV) model to predict the malicious nodes in the WSN. The fitness functions of the HVAB-NIV have operated to recognize the energy level of each node and improve the performance of node detection. The developed replica includes three stages that monitor each node, calculate the energy level and detect the malicious node. More than 100 node inputs were initialized in the proposed technique and implemented in the MATLAB tool. The suggested mechanism enhances the performance of malicious node detection and gains good accuracy for detecting nodes also, it saves running time and power consumption. The experimental results of the developed model has validated with other existing replicas to running time, False Prediction Rate (FPR), detection accuracy, True Prediction Rate (TPR), and power consumption. The developed methods achieve better results by gaining a high rate of accuracy detection, less running time, and false rate detection.

Wireless sensor networks (WSNs) have emerged as a significant architecture for data collection in various applications. However, the integration of WSNs with IoT poses energy-related challenges due to limited sensor node energy, increased energy consumption for wireless data sharing, and the necessity of energy-efficient routing protocols for reliable transmission and reduced energy consumption. This paper proposes an optimized energy-efficient routing protocol for wireless sensor networks integrated with the Internet of Things. The protocol aims to improve network lifetime and secure data transmission by identifying the optimal Cluster Heads (CHs) in the network, selected using a Tree Hierarchical Deep Convolutional Neural Network. To achieve this, the paper introduces a fitness function that takes into account cluster density, traffic rate, energy, collision, delay throughput, and distance from the capacity node. Additionally, the paper considers three factors, including trust, connectivity, and QoS, to determine the best course of action. The paper also presents a novel optimization approach, using the hybrid Marine Predators Algorithm (MPA) and Woodpecker Mating Algorithm (WMA), to optimize trust, connectivity, and QoS parameters for optimal path selection with minimal delay. The simulation process is implemented in MATLAB, and the developed method’s efficiency is evaluated using several performance metrics. The results of the simulation demonstrate the effectiveness of the proposed method, which achieved significantly lower delay (99.67%, 98.38%, 89.34%, and 97.45%), higher delivery ratio (89.34%, 89.34%, 83.12%, and 88.96%), and lower packet drop (93.15%, 91.25%, 79.90%, and 92.88%) in comparison to existing methods. These outcomes indicate the potential of the optimized energy-efficient routing protocol to improve network lifetime and ensure secure data transmission in WSNs integrated with IoT.