Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Battery-aware Cyclic Scheduling in Energy-harvesting Federated Learning

Eunjeong Jeong, Nikolaos Pappas

TL;DR

The paper addresses the energy challenges of Federated Learning in energy-harvesting environments by proposing FedBacys, a battery-aware cyclic participation framework. FedBacys clusters users by remaining energy and enforces sequential intra-group updates across epochs, effectively reducing redundant local training and balancing energy usage while considering both computation and communication costs. Through CIFAR-10 experiments, it demonstrates faster convergence and lower energy consumption across varying battery- charging scenarios and group counts, with robustness to non-iid data distributions. The work provides a practical scheduling strategy for sustainable EHFL deployments on energy-constrained devices, and outlines future directions for semantics-aware participation and hyperparameter interdependencies.

Abstract

Federated Learning (FL) has emerged as a promising framework for distributed learning, but its growing complexity has led to significant energy consumption, particularly from computations on the client side. This challenge is especially critical in energy-harvesting FL (EHFL) systems, where device availability fluctuates due to limited and time-varying energy resources. We propose FedBacys, a battery-aware FL framework that introduces cyclic client participation based on users' battery levels to cope with these issues. FedBacys enables clients to save energy and strategically perform local training just before their designated transmission time by clustering clients and scheduling their involvement sequentially. This design minimizes redundant computation, reduces system-wide energy usage, and improves learning stability. Our experiments demonstrate that FedBacys outperforms existing approaches in terms of energy efficiency and performance consistency, exhibiting robustness even under non-i.i.d. training data distributions and with very infrequent battery charging. This work presents the first comprehensive evaluation of cyclic client participation in EHFL, incorporating both communication and computation costs into a unified, resource-aware scheduling strategy.

Battery-aware Cyclic Scheduling in Energy-harvesting Federated Learning

TL;DR

The paper addresses the energy challenges of Federated Learning in energy-harvesting environments by proposing FedBacys, a battery-aware cyclic participation framework. FedBacys clusters users by remaining energy and enforces sequential intra-group updates across epochs, effectively reducing redundant local training and balancing energy usage while considering both computation and communication costs. Through CIFAR-10 experiments, it demonstrates faster convergence and lower energy consumption across varying battery- charging scenarios and group counts, with robustness to non-iid data distributions. The work provides a practical scheduling strategy for sustainable EHFL deployments on energy-constrained devices, and outlines future directions for semantics-aware participation and hyperparameter interdependencies.

Abstract

Federated Learning (FL) has emerged as a promising framework for distributed learning, but its growing complexity has led to significant energy consumption, particularly from computations on the client side. This challenge is especially critical in energy-harvesting FL (EHFL) systems, where device availability fluctuates due to limited and time-varying energy resources. We propose FedBacys, a battery-aware FL framework that introduces cyclic client participation based on users' battery levels to cope with these issues. FedBacys enables clients to save energy and strategically perform local training just before their designated transmission time by clustering clients and scheduling their involvement sequentially. This design minimizes redundant computation, reduces system-wide energy usage, and improves learning stability. Our experiments demonstrate that FedBacys outperforms existing approaches in terms of energy efficiency and performance consistency, exhibiting robustness even under non-i.i.d. training data distributions and with very infrequent battery charging. This work presents the first comprehensive evaluation of cyclic client participation in EHFL, incorporating both communication and computation costs into a unified, resource-aware scheduling strategy.

Paper Structure

This paper contains 5 sections, 4 equations, 5 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. System Model
  3. Energy-harvesting Cyclic FL
  4. Experimental Results
  5. Conclusions and Future Directions

Figures (5)

  • Figure 1: A schematic view of energy-harvesting FL based on probabilistic decision rules.
  • Figure 2: A simplified example of a FedBacys network with $N=10$ users assigned into $G=3$ groups. Within each group, (1) each user transmits local updates to the group's hub if they satisfy the conditions to send local updates; (2) the hub aggregates the received updates; (3) the hub sends the updated intra-group model to the next group by multicast. The final group's hub sends the aggregated model to the server, which is then sent to Group 1 of the new epoch. The hub users can be changed at the beginning of every epoch.
  • Figure 3: Test accuracy w.r.t. different algorithms ($G=10$ groups)
  • Figure 4: Average test accuracy of FedBacys w.r.t. different battery charging probabilities. ($G=2,\ 5$)
  • Figure 5: Average test accuracy of FedBacys on non-i.i.d. training datasets with different Dirichlet coefficients applied. $p_{bc}$ is fixed to 0.5 for all trials.