TY  - JOUR
AU  - Nahak, Manoswini
AU  - Jana, Atanu
AU  - Panda, Snehal
AU  - Singaravel, G.
AU  - Mallik, Sandipan
AU  - Swain, Kunjabihari
PY  - 2026
DA  - 2026/06/26
TI  - Energy-Efficient AIoT Solutions: A Critical Review of Power Consumption Models, Machine Learning-Based Energy Optimization, and Deployment Strategies for Sustainable IoT Networks
JO  - Next-Generation Computing Systems and Technologies
T2  - Next-Generation Computing Systems and Technologies
JF  - Next-Generation Computing Systems and Technologies
VL  - 2
IS  - 2
SP  - 51
EP  - 58
DO  - 10.62762/NGCST.2026.276719
UR  - https://www.icck.org/article/abs/NGCST.2026.276719
KW  - AIoT
KW  - energy efficiency
KW  - power consumption modeling
KW  - TinyML
KW  - edge computing
KW  - sustainable IoT
AB  - The integration of artificial intelligence and the Internet of Things (AIoT) enables advanced edge computing but creates significant energy challenges for resource-constrained devices. To address these challenges, this paper introduces a novel comparative taxonomy and a systematic gap analysis matrix that directly maps hardware-aware TinyML paradigms to network-layer scheduling in sustainable AIoT systems. Synthesizing recent empirical studies (2023–2026) on power consumption models for IoT edge devices, machine learning techniques for energy optimization, and sustainable deployment strategies, we demonstrate that additive and regression-based models achieve low prediction error (MAPE 4–8%) on single-board computers. Furthermore, hybrid deep learning and reinforcement learning approaches, combined with model compression and hierarchical edge-fog-cloud architectures, achieve 20–45% energy savings. Key challenges identified include hardware heterogeneity, training overhead, and the critical performance gap between laboratory environments and real-world deployments. Finally, future research directions focus on hardware-aware TinyML frameworks, federated learning paradigms, and energy-carbon co-optimization metrics to establish truly sustainable IoT networks.
SN  - 3070-3328
PB  - Institute of Central Computation and Knowledge
LA  - English
ER  - 

@article{Nahak2026EnergyEffi,
  author = {Manoswini Nahak and Atanu Jana and Snehal Panda and G. Singaravel and Sandipan Mallik and Kunjabihari Swain},
  title = {Energy-Efficient AIoT Solutions: A Critical Review of Power Consumption Models, Machine Learning-Based Energy Optimization, and Deployment Strategies for Sustainable IoT Networks},
  journal = {Next-Generation Computing Systems and Technologies},
  year = {2026},
  volume = {2},
  number = {2},
  pages = {51-58},
  doi = {10.62762/NGCST.2026.276719},
  url = {https://www.icck.org/article/abs/NGCST.2026.276719},
  abstract = {The integration of artificial intelligence and the Internet of Things (AIoT) enables advanced edge computing but creates significant energy challenges for resource-constrained devices. To address these challenges, this paper introduces a novel comparative taxonomy and a systematic gap analysis matrix that directly maps hardware-aware TinyML paradigms to network-layer scheduling in sustainable AIoT systems. Synthesizing recent empirical studies (2023–2026) on power consumption models for IoT edge devices, machine learning techniques for energy optimization, and sustainable deployment strategies, we demonstrate that additive and regression-based models achieve low prediction error (MAPE 4–8\%) on single-board computers. Furthermore, hybrid deep learning and reinforcement learning approaches, combined with model compression and hierarchical edge-fog-cloud architectures, achieve 20–45\% energy savings. Key challenges identified include hardware heterogeneity, training overhead, and the critical performance gap between laboratory environments and real-world deployments. Finally, future research directions focus on hardware-aware TinyML frameworks, federated learning paradigms, and energy-carbon co-optimization metrics to establish truly sustainable IoT networks.},
  keywords = {AIoT, energy efficiency, power consumption modeling, TinyML, edge computing, sustainable IoT},
  issn = {3070-3328},
  publisher = {Institute of Central Computation and Knowledge}
}