Abstract
In recent years, networks of sensors have gained significant attention for security-sensitive applications, such as border monitoring, where network durability, efficiency, and security are of utmost importance. In sensor networks, security related measure will often lead to an increase in energy consumption, hence, designing efficient energy conserving protocols with robust data aggregation methods are crucial in line with better duty-cycling idle nodes in the network. This paper presents a Dual Layer Sea Lion Optimization algorithm (DL-SLnOA) model, a bio-inspired, clustering and routing that focuses on energy efficiency and security to enhance performance under challenging conditions. The proposed DL-SLnOA combines the SLnO algorithm with a dual-layer security framework to optimize the selection of Cluster Heads (CHs) by using adaptive exploration and exploitation methods, which dynamically position CHs to balance energy consumption, proximity, and trustworthiness. DL-SLnOA first layer incorporates dynamic trust score update to ensure only reliable nodes participate in communication and second layer implements anomaly detection methods, which are used to identify malicious behaviour with inherent high detection rates and least false positive rates. Unlike other methods, this solution tends to firmly segregate legitimate from malicious nodes, thereby reinforcing the network. The simulation results for selective forwarding attacks with 10 malicious nodes in 100 iterations detected threats within 7 rounds with a packet transmission efficiency of 97.9%, while for wormhole attacks detection within 2 rounds, packet transmission efficiency of 98.3%, and average energy consumption of about 0.122J, significantly increasing the network lifetime. These improvements that provide an energy-efficient and secure platform strongly reinforce the network against security threats by further extending the life span of the network, making DL-SLnOA a scalable solution.