Research Article

Occupancy Inference from Aggregate Smart-Meter Data Across Sampling Resolutions: A Controlled Synthetic Privacy Study

Authors

Zeinab Parandavar * Department of Computer Engineering, Rasht Branch, Islamic Azad University, Rasht, Iran.
Arman Daliri Department of Computer Engineering, Karaj Branch, Islamic Azad University, Karaj, Iran.

https://doi.org/10.48314/ceti.v3i3.95

Downloads

Abstract

Fine-grained smart-meter records support legitimate energy services while also enabling behavioral inference. This study isolates one controllable disclosure parameter—the reporting interval—and examines how it changes retrospective occupancy inference in a controlled synthetic benchmark. Sixty households are simulated for 14 days at one-second native resolution, then mean-aggregated to 1-second, 1-minute, 5-minute, and 15-minute releases. Non-overlapping 30-minute windows are represented by statistical and temporal features and evaluated with logistic regression, random forest, gradient boosting, and a two-state Gaussian Hidden Markov Model (HMM). The primary protocol trains on 42 households and evaluates 18 previously unseen households (12,096 windows); uncertainty is summarized by a 500-draw household-cluster bootstrap. Random forest attains its highest pooled Receiver Operating Characteristic – Area Under the Curve (ROC AUC) at 1 minute (0.794; 95% bootstrap interval 0.768–0.817) and declines to 0.724 (0.703–0.747) at 15 minutes. Gradient boosting reaches 0.734 (0.708–0.761) at 15 minutes. Statistical-only feature sets consistently outperform temporal-only sets, showing that coarse releases can retain occupancy-linked level information after short transitions are suppressed. The HMM remains at AUC 0.722 at every resolution because it receives only the 30-minute mean, which is invariant under the nested averaging scheme. The analysis also exposes three interpretive boundaries: The midpoint label is estimated using the complete window and is therefore retrospective; a secondary time split changes several design factors and cannot identify the causal value of household-specific history; and the AUC 0.650 line is an analytical benchmark rather than a legal or operational privacy threshold. The findings support data minimization as a useful but incomplete control: 15-minute averaging reduces some model performance without eliminating inferential exposure in this generator. External validation on occupancy-labeled homes, prospective inference, and release of code, generated traces, predictions, and bootstrap draws are required before deployment claims.

Keywords

Smart meter Occupancy inference Data granularity Privacy exposure Synthetic simulation Household-cluster bootstrap

References (30)

[1] Hart, G. W. (1992). Nonintrusive appliance load monitoring. Proceedings of the IEEE (pp. 1870–1891). IEEE. https://doi.org/10.1109/5.192069

[2] Molina-Markham, A., Shenoy, P., Fu, K., Cecchet, E., & Irwin, D. (2010). Private memoirs of a smart meters. Proceedings of the 2nd ACM workshop on embedded sensing systems for energy-efficiency in building (pp. 61–66). Association for Computing Machinery (ACM). https://doi.org/10.1145/1878431.1878446

[3] Lisovich, M. A., Mulligan, D. K., & Wicker, S. B. (2010). Inferring personal information from demand-response systems. IEEE Security & Privacy, 8(1), 11–20. https://doi.org/10.1109/MSP.2010.40

[4] McKenna, E., Richardson, I., & Thomson, M. (2012). Smart meter data: Balancing consumer privacy concerns with legitimate applications. Energy Policy, 41, 807–814. https://doi.org/10.1016/j.enpol.2011.11.049

[5] Beckel, C., Sadamori, L., Staake, T., & Santini, S. (2014). Revealing household characteristics from smart meter data. Energy, 78, 397–410. https://doi.org/10.1016/j.energy.2014.10.025

[6] Chen, D., Barker, S., Subbaswamy, A., Irwin, D., & Shenoy, P. (2013). Non-intrusive occupancy monitoring using smart meters. Proceedings of the 5th ACM workshop on embedded systems for energy-efficient buildings (pp. 1–8). Association for Computing Machinery (ACM). https://doi.org/10.1145/2528282.2528294

[7] Kleiminger, W., Beckel, C., Staake, T., & Santini, S. (2013). Occupancy detection from electricity consumption data. Proceedings of the 5th ACM workshop on embedded systems for energy-efficient buildings (pp. 1–8). Association for Computing Machinery (ACM). https://doi.org/10.1145/2528282.2528295

[8] Kleiminger, W., Beckel, C., & Santini, S. (2015). Household occupancy monitoring using electricity meters. Proceedings of the 2015 ACM international joint conference on pervasive and ubiquitous computing (pp. 975–986). Association for Computing Machinery (ACM). https://doi.org/10.1145/2750858.2807538

[9] Tang, G., Wu, K., Lei, J., & Xiao, W. (2015). The meter tells you are at home! non-intrusive occupancy detection via load curve data. 2015 IEEE international conference on smart grid communications (smartgridcomm) (pp. 897–902). IEEE. https://doi.org/10.1109/SmartGridComm.2015.7436415

[10] Razavi, R., Gharipour, A., Fleury, M., & Akpan, I. J. (2019). Occupancy detection of residential buildings using smart meter data: A large-scale study. Energy and Buildings, 183, 195–208. https://doi.org/10.1016/j.enbuild.2018.11.025

[11] Eibl, G., & Engel, D. (2014). Influence of data granularity on smart meter privacy. IEEE Transactions on Smart Grid, 6(2), 930–939. https://doi.org/10.1109/TSG.2014.2376613

[12] Hernandez, J. C., Sanchez-Sutil, F., Cano-Ortega, A., & Baier, C. R. (2020). Influence of data sampling frequency on household consumption load profile features: A case study in Spain. Sensors, 20(21), 6034. https://doi.org/10.3390/s20216034

[13] Radovanovic, D., Unterweger, A., Eibl, G., Engel, D., & Reichl, J. (2022). How unique is weekly smart meter data? Energy Informatics, 5(Suppl 1), 13. https://doi.org/10.1186/s42162-022-00205-8

[14] Farokhi, F. (2020). Review of results on smart-meter privacy by data manipulation, demand shaping, and load scheduling. IET Smart Grid, 3(5), 605–613. https://doi.org/10.1049/iet-stg.2020.0129

[15] Giaconi, G., Gunduz, D., & Poor, H. V. (2018). Privacy-aware smart metering: Progress and challenges. IEEE Signal Processing Magazine, 35(6), 59–78. https://doi.org/10.1109/MSP.2018.2841410

[16] Giaconi, G., Gündüz, D., & Poor, H. V. (2021). 10 smart meter data privacy. Advanced data analytics for power systems (pp. 230–260). Cambridge University Press. https://doi.org/10.1017/9781108859806.014

[17] Kua, J., Hossain, M. B., Natgunanathan, I., & Xiang, Y. (2023). Privacy preservation in smart meters: Current status, challenges and future directions. Sensors, 23(7), 3697. https://doi.org/10.3390/s23073697

[18] Abdalzaher, M. S., Fouda, M. M., & Ibrahem, M. I. (2022). Data privacy preservation and security in smart metering systems. Energies, 15(19), 7419. https://doi.org/10.3390/en15197419

[19] Eibl, G., & Engel, D. (2017). Differential privacy for real smart metering data. Computer Science-Research and Development, 32(1), 173–182. https://doi.org/10.1007/s00450-016-0310-y

[20] Richardson, I., Thomson, M., & Infield, D. (2008). A high-resolution domestic building occupancy model for energy demand simulations. Energy and Buildings, 40(8), 1560–1566. https://doi.org/10.1016/j.enbuild.2008.02.006

[21] Richardson, I., Thomson, M., Infield, D., & Clifford, C. (2010). Domestic electricity use: A high-resolution energy demand model. Energy and Buildings, 42(10), 1878–1887. https://doi.org/10.1016/j.enbuild.2010.05.023

[22] Widén, J., & Wäckelgård, E. (2010). A high-resolution stochastic model of domestic activity patterns and electricity demand. Applied Energy, 87(6), 1880–1892. https://doi.org/10.1016/j.apenergy.2009.11.006

[23] Sayed, A. N., Himeur, Y., & Bensaali, F. (2022). Deep and transfer learning for building occupancy detection: A review and comparative analysis. Engineering Applications of Artificial Intelligence, 115, 105254. https://doi.org/10.1016/j.engappai.2022.105254

[24] Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., ..., & Duchesnay, É. (2011). Scikit-learn: Machine learning in Python. Journal of Machine Learning Research, 12, 2825–2830. https://jmlr.org/papers/v12/pedregosa11a.html

[25] Breiman, L. (2001). Random forests. Machine learning, 45(1), 5–32. https://doi.org/10.1023/A:1010933404324

[26] Friedman, J. H. (2001). Greedy function approximation: A gradient boosting machine. Annals of statistics, 29(5), 1189–1232. https://doi.org/10.1214/aos/1013203451

[27] Rabiner, L. R. (1989). A tutorial on hidden Markov models and selected applications in speech recognition. Proceedings of the IEEE (pp. 257–286). IEEE. https://doi.org/10.1109/5.18626

[28] Efron, B. (1997). Bootstrap methods: Another look at the Jackknife. Annals of Statistics, 7(1), 1–26. https://doi.org/10.1214/aos/1176344552

[29] Friedman, M. (1937). The use of ranks to avoid the assumption of normality implicit in the analysis of variance. Journal of the American Statistical Association, 32(200), 675–701. https://doi.org/10.1080/01621459.1937.10503522

[30] Wilcoxon, F. (1945). Individual comparisons by ranking methods. Biometrics Bulletin, 1(6), 80–83. https://doi.org/10.2307/3001968

Article Metrics

Published 2026-09-18
Issue Vol. 3 No. 3 (2026): Computational Engineering and Technology Innovations
Section Articles
Document Type Research Article
Views 0
Downloads 0

How to Cite

Parandavar, Z., & Daliri, A. (2026). Occupancy Inference from Aggregate Smart-Meter Data Across Sampling Resolutions: A Controlled Synthetic Privacy Study. Computational Engineering and Technology Innovations, 3(3), 249-264. https://doi.org/10.48314/ceti.v3i3.95

Download Citation

Similar Articles

Hashem Saberi Najafi, Szabolcs Fischer, Isametova Madina Esdauletova, Deep Reinforcement Learning-Enhanced Levenberg-Marquardt Neural Network for Improved Energy Efficiency in Wireless Sensor Networks , Computational Engineering and Technology Innovations: Vol. 1 No. 3 (2024): Computational Engineering and Technology Innovations
Victoria Nozick, Fatemeh Rasoulpour, Milos Milovancevic, View Coverage and Evidence Gating for Vision-Based Robotic Quality Inspection , Computational Engineering and Technology Innovations: Vol. 3 No. 3 (2026): Computational Engineering and Technology Innovations
Omar Mar Cornelio, Mingyue Wang, Rania A. Mahmoud, Exact Predictive Ventilation Control with Causal Occupancy Forecasts and Concentration Guarantees , Computational Engineering and Technology Innovations: Vol. 3 No. 2 (2026): Computational Engineering and Technology Innovations
Javad Pourqasem *, Markus Hess, Home Appliance Control and Power Monitoring Using I2C Communication , Computational Engineering and Technology Innovations: Vol. 1 No. 1 (2024): Computational Engineering and Technology Innovations
Renas Rajab Asaad, Madasu Hanmandlu, Kharabela Swain, AISR: A Protocol-Closed Architecture for Interference-Aware Channel-Mask Control in Industrial LoRaWAN , Computational Engineering and Technology Innovations: Vol. 3 No. 2 (2026): Computational Engineering and Technology Innovations
Amrit Raj, Fog Computing in Smart City IoT Architectures for Latency Reduction , Computational Engineering and Technology Innovations: Vol. 1 No. 4 (2024): Computational Engineering and Technology Innovations
Abhisek Mishra, Edge Computing and IoT-Based Urban Environment Monitoring Solutions , Computational Engineering and Technology Innovations: Vol. 1 No. 3 (2024): Computational Engineering and Technology Innovations
Shivendra Dubey, Cloud-Edge Architecture for IoT in Smart Building Automation , Computational Engineering and Technology Innovations: Vol. 1 No. 3 (2024): Computational Engineering and Technology Innovations
Renas Rajab Asaad, Viktor P. Sokolov, Adaptive Accuracy and Cost Control under Distribution Shift in Model Cascades , Computational Engineering and Technology Innovations: Vol. 3 No. 3 (2026): Computational Engineering and Technology Innovations
Pragati Das, AI-Based Network Management for IoT Devices , Computational Engineering and Technology Innovations: Vol. 1 No. 1 (2024): Computational Engineering and Technology Innovations

1-10 of 60 Next

You may also start an advanced similarity search for this article.