Light weight Intelligence-TinyML based Energy Usage Prediction for Resource-Constrained IoT Edges Nodes

Sandeep Kumar Rawat; Neha Tuli

doi:10.69968/ijisem.2026v5i3247-253

Authors

Sandeep Kumar Rawat Research Scholar, Sri Sai University, Palampur, Himachal Pradesh, India,
Neha Tuli Department of CSE, Sri Sai University, Palampur, Himachal Pradesh, India

DOI:

https://doi.org/10.69968/ijisem.2026v5i3247-253

Keywords:

TinyML, Internet of Things, Energy Prediction, Edge Computing, Low-Power Systems, Energy-Efficient IoT

Abstract

The extensive deployment of battery-powered and resource-constrained edge devices makes energy efficiency a major challenge in Internet of Things (IoT) systems. Accurate energy prediction is important to enable intelligent energy management. However, traditional machine learning models are usually computationally expensive and unsuitable for micro-controller based platforms. In this paper, we present a TinyML-based energy prediction framework for low-power IoT edge devices. The proposed approach employs lightweight machine learning models which are optimized for ultra-low memory and computation footprints, but still retain acceptable prediction accuracy. We collect energy consumption data from a real IoT testbed, and train and evaluate several TinyML compatible models. The experimental results show that the proposed TinyML-based predictor can provide reliable energy estimation with low inference latency and low memory overhead, and thus can be deployed on resource-constrained IoT devices. This work lays a fundamental foundation for intelligent and adaptive energy management in future IoT systems.

References

[1] J. Polastre et al., “Telos: Enabling ultra-low power wireless research,” ACM IPSN, 2005.

[2] D. Culler et al., “Wireless sensor networks,” Communications of the ACM, 2004.

[3] R. Min et al., “Energy-centric enabling technologies for wireless sensor networks,” IEEE Wireless Communications, 2002.

[4] L. Zhang et al., “Energy consumption modeling for embedded systems,” IEEE Transactions on Computers, 2010.

[5] S. Rivoire et al., “Energy complexity of algorithms,” ACM Journal on Emerging Technologies, 2007.

[6] Kansal et al., “Power management in embedded systems,” IEEE Computer, 2007.

[7] Y. Kim et al., “Machine learning based energy prediction,” IEEE Embedded Systems Letters, 2013.

[8] S. Singh et al., “Decision tree-based energy modeling,” Sustainable Computing, 2016.

[9] M. Chen et al., “Neural network-based energy estimation,” IEEE Access, 2018.

[10] X. Xu et al., “Cloud-assisted IoT energy optimization,” Future Generation Computer Systems, 2019.

[11] H. Gupta et al., “Cloud-based energy analytics for IoT,” IEEE IoT Journal, 2020.

[12] W. Shi et al., “Edge computing: Vision and challenges,” IEEE Internet of Things Journal, 2016.

[13] Y. Mao et al., “A survey on mobile edge computing,” IEEE Communications Surveys, 2017.

[14] T. Han et al., “Energy-aware task scheduling in IoT,” IEEE Systems Journal, 2018.

[15] J. Xu et al., “Predictive energy management in IoT,” Ad Hoc Networks, 2019.

[16] P. Warden and D. Situnayake, TinyML, O’Reilly Media, 2020.

[17] N. D. Lane et al., “An early resource characterization of deep learning on wearables,” ACM WearSys, 2015.

[18] S. Banbury et al., “Benchmarking TinyML systems,” NeurIPS TinyML Workshop, 2021.

[19] F. Conti et al., “Efficient inference on microcontrollers,” IEEE Micro, 2017.

[20] Capotondi et al., “CMix-NN: Mixed low-precision CNN library,” IEEE Access, 2020.

[21] Mathur et al., “TinyML for embedded intelligence,” IEEE Design & Test, 2021.

[22] D. G. Costa et al., “Lightweight ML at the edge,” Sensors, 2022.

[23] K. Raghavan et al., “Energy estimation in wireless sensor networks,” Journal of Network and Computer Applications, 2018.

[24] S. Liu et al., “Low-power energy modeling for embedded devices,” Microprocessors and Microsystems, 2019.

[25] M. S. Hossain et al., “Energy-efficient IoT systems,” IEEE Access, 2021.

[26] Banbury, C. R., et al., “Benchmarking TinyML systems: Energy, latency, and memory trade-offs,” IEEE Internet of Things Journal, vol. 11, no. 2, pp. 1450–1464, 2024.

[27] Hussein, A., Ahmed, M., and Kim, J., “Energy-aware TinyML inference on microcontroller-based IoT devices,” IEEE Access, vol. 12, pp. 38721–38735, 2024.

[28] Zhang, Y., Li, X., and Wang, H., “Lightweight machine learning models for energy prediction in edge IoT systems,” Future Generation Computer Systems, vol. 148, pp. 215–227, 2024.

[29] Costa, D. G., Guedes, L. A., and Vasques, F., “TinyML-based edge intelligence for sustainable IoT deployments,” Sensors, vol. 24, no. 3, Art. no. 1125, 2024.

[30] Rahman, M., Hossain, M. S., and Muhammad, G., “Edge intelligence for energy-efficient Internet of Things: A TinyML perspective,” IEEE Communications Surveys & Tutorials, vol. 26, no. 1, pp. 410–438, 2024.

[31] Singh, R., Verma, P., and Kumar, N., “On-device energy consumption prediction using quantized neural networks for IoT nodes,” Ad Hoc Networks, vol. 154, Art. no. 103351, 2025.

[32] Al-Fuqaha, A., et al., “TinyML-enabled edge analytics for resource-constrained IoT systems,” IEEE Internet of Things Journal, Early Access, 2025.