An Optimal-Aware Energy Efficient Framework for Energy-Constrained Networks in the Internet of Things

Abstract

The Internet of Things (IoT) refers to a rapidly expanding network of interconnected devices designed to provide intelligent and context-aware services across various industries, including smart cities, agriculture, healthcare, and industrial automation. This paper studies the creation of an energy-efficient routing framework in the context of the IoT that takes Quality of Service (QoS) parameters into account since existing protocols fail to consider important QoS metrics like Residual Energy (RE), Packet Delivery Ratio (PDR), End-to-End Delay (E2E), latency, throughput, and reliability, amongst others. They aim to improve communication reliability and extend the network lifetime in dynamic IoT scenarios. The proposed technique and framework integrate a QoS-accommodating behaviour in energy-efficient routing, following a multi-criteria decision mechanism based on RE, trust level, Hop Count (HC), node load, and delay. The path selection approach is based on the clustering protocol Threshold Distributed Energy-Efficient Clustering (TDEEC). Still, it incorporates improved performance using metaheuristic algorithms, specifically the hybrid Grey Wolf Optimization and Ant Colony Optimization (GWO-ACO). Challenges, such as the dynamic selection of the Cluster Heads (CHs), adaptively choosing the secure routing path to improve the network for specific QoS requirements relevant to the application, are also addressed. A large number of simulations tests the performance of the proposed technique. The QoS-aware routing model yields noticeable improvements in various performance metrics. It can achieve a PDR of 98.73%, energy savings of 10.81%, and a 10.92% longer network lifetime compared to conventional routing methods. Additionally, it performs well even under variable network conditions, offering improved latency, throughput, reliability, and RE. This paper proposes a new, scalable, and secure routing approach for IoT networks that leverages QoS awareness, energy-efficient clustering, and intelligent path selection through hybrid metaheuristic algorithms. In contrast to conventional methods, this solution can adapt to changes in the network state in a way that ensures the QoS required is still guaranteed, making it a relevant solution for time-sensitive, large-scale IoT applications.