Deep Learning-based Data Collection to Reduce Energy Consumption in Internet of Things Devices
Lorenzo Calisti, Emanuele Lattanzi · 2025
The Internet of Things has revolutionized various industries, but the energy consumption of the devices remains a significant challenge. This paper proposes a novel approach to reduce energy consumption using deep learning-based data collection techniques. Our proposed framework leverages deep learning models to forecast sensor data to optimize data transmission to adapt to dynamic environmental conditions and reduce unnecessary energy expenditure. We demonstrate the effectiveness of our approach through extensive experiments, showcasing energy savings of up to 93% with respect to the state-of-the-art methods while maintaining desired levels of data accuracy and system performance. This research contributes to the development of more energy-efficient and sustainable Internet of Things systems, paving the way for a greener future.