Enhancing consumer privacy in federated load forecasting through single layer aggregation

Habib Ullah Manzoor*, Muhammad Ali Jamshed, Ahmed Zoha, Sanaullah Manzoor

*Corresponding author for this work

Research output: Contribution to specialist publicationArticle

20 Downloads (Pure)

Abstract

Consumer energy data presents unique challenges in terms of privacy and security due to the vast amounts generated by IoT-based devices. Federated learning (FL) has emerged as a privacy-preserving and cost-efficient approach for training machine learning models, offering solutions to protect consumers' data from breaches. However, FL models are still vulnerable to privacy attacks. To mitigate these risks, differential privacy (DP) is applied to local model weights, enhancing privacy but often reducing prediction accuracy. To address this trade-off, we propose a novel single-layer federated weight aggregation framework for consumer energy load forecasting. This privacy-aware approach aggregates only the single layer of local models, ignoring other layers to minimize privacy risks. Extensive simulations with real-world consumer energy data show the framework's effectiveness, achieving a Mean Absolute Error (MAE) of 1.13kWh, close to the baseline of 1.01kWh. It significantly outperforms traditional FL-based methods (FedAvg), which have an MAE of 4.2kWh at a noise level (ϵ) of 0.2, while providing 211.09 times better communication cost efficiency.

Original languageEnglish
Specialist publicationIEEE Consumer Electronics Magazine
PublisherInstitute of Electrical and Electronics Engineers
DOIs
Publication statusE-pub ahead of print - 18 Nov 2024

ASJC Scopus subject areas

  • Human-Computer Interaction
  • Hardware and Architecture
  • Computer Science Applications
  • Electrical and Electronic Engineering

Fingerprint

Dive into the research topics of 'Enhancing consumer privacy in federated load forecasting through single layer aggregation'. Together they form a unique fingerprint.

Cite this