OPTIMIZATION OF COMMUNICATION SYSTEMS FOR ENERGY EFFICIENCY USING MACHINE LEARNING TECHNIQUES

Abstract

The rapid expansion of wireless communication networks and connected devices has significantly increased energy consumption across network infrastructures. Energy efficiency has therefore become a critical design objective for modern and next-generation communication systems. This paper presents energy-efficient communication architectures that leverage machine learning techniques to optimize resource utilization and reduce power consumption. The proposed approach applies learning-based models to dynamically manage transmission power, scheduling, and spectrum allocation under varying traffic and channel conditions. By continuously adapting network parameters, the architecture minimizes unnecessary energy expenditure while maintaining quality-of-service requirements. Simulation results demonstrate notable improvements in energy efficiency compared to conventional static and rule-based communication systems. The findings highlight the potential of machine learning–driven architectures to support sustainable and green wireless communications.