Adaptive Multi-Objective Optimization Framework for 5G-Enabled IoT Networks: Enhancing Performance and Energy Efficiency in Smart Application Environments

Abstract

The convergence of fifth generation (5G) networks and the Internet of Things (IoT) is composed to revolutionize smart application environments, such as smart cities and industrial automation. However, this integration introduces significant challenges, primarily the conflicting demands of maximizing network performance (e.g., throughput, latency) and minimizing energy consumption, which is projected to surge with hyper-dense deployments. This paper proposes a novel Adaptive Multi-Objective Optimization (AMOO) framework designed to address this trade-off in 5G-enabled IoT networks. The framework leverages machine learning to dynamically adapt optimization strategies based on real-time network conditions and traffic patterns. We formulate a multi-objective problem to simultaneously optimize spectral efficiency (SE) and energy efficiency (EE). The core of our framework is an AI-enhanced Non-Dominated Sorting Genetic Algorithm (AI-NSGA), which intelligently explores the solution space for resource allocation, including power control and small cell activation. Simulation results, modeled in a MATLAB environment, demonstrate that our proposed AMOO framework achieves a superior Pareto front compared to static optimization and single-objective approaches. It goes along with substantial energy savings up to 25% in our scenarios while maintaining high Quality of Service (QoS) levels, providing a scalable and efficient solution for sustainable smart city and Industrial IoT (IIoT) deployments.