Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Joint Time and Energy-Efficient Federated Learning-based Computation Offloading Method for Mobile Edge Computing

Anwesha Mukherjee, Rajkumar Buyya

TL;DR

The paper tackles energy- and latency-sensitive computation offloading in mobile edge computing by introducing two federated learning–based methods. FLDec performs offloading decisions by first classifying task intensity with an MLP and then deciding offload vs local execution with an LSTM, achieving high prediction accuracy in real-time settings. FedOff enables secure partial offloading by partitioning local data between device-side processing and edge-server training, incorporating symmetric-key encryption to protect transmissions while preserving high global/local model accuracy. Experiments demonstrate substantial reductions in response time and device energy (11–31% for intensive tasks) and minimal encryption overhead (0.05–0.16%), with FedOff reaching global model accuracy >98% and local accuracy >94% across tasks. The work advances privacy-preserving, time- and energy-efficient MEC offloading, with practical implications for IoT and mobile applications.

Abstract

Computation offloading at lower time and lower energy consumption is crucial for resource limited mobile devices. This paper proposes an offloading decision-making model using federated learning. Based on the task type and the user input, the proposed decision-making model predicts whether the task is computationally intensive or not. If the predicted result is computationally intensive, then based on the network parameters the proposed decision-making model predicts whether to offload or locally execute the task. According to the predicted result the task is either locally executed or offloaded to the edge server. The proposed method is implemented in a real-time environment, and the experimental results show that the proposed method has achieved above 90% prediction accuracy in offloading decision-making. The experimental results also present that the proposed offloading method reduces the response time and energy consumption of the user device by ~11-31% for computationally intensive tasks. A partial computation offloading method for federated learning is also proposed and implemented in this paper, where the devices which are unable to analyse the huge number of data samples, offload a part of their local datasets to the edge server. For secure data transmission, cryptography is used. The experimental results present that using encryption and decryption the total time is increased by only 0.05-0.16%. The results also present that the proposed partial computation offloading method for federated learning has achieved a prediction accuracy of above 98% for the global model.

A Joint Time and Energy-Efficient Federated Learning-based Computation Offloading Method for Mobile Edge Computing

TL;DR

The paper tackles energy- and latency-sensitive computation offloading in mobile edge computing by introducing two federated learning–based methods. FLDec performs offloading decisions by first classifying task intensity with an MLP and then deciding offload vs local execution with an LSTM, achieving high prediction accuracy in real-time settings. FedOff enables secure partial offloading by partitioning local data between device-side processing and edge-server training, incorporating symmetric-key encryption to protect transmissions while preserving high global/local model accuracy. Experiments demonstrate substantial reductions in response time and device energy (11–31% for intensive tasks) and minimal encryption overhead (0.05–0.16%), with FedOff reaching global model accuracy >98% and local accuracy >94% across tasks. The work advances privacy-preserving, time- and energy-efficient MEC offloading, with practical implications for IoT and mobile applications.

Abstract

Computation offloading at lower time and lower energy consumption is crucial for resource limited mobile devices. This paper proposes an offloading decision-making model using federated learning. Based on the task type and the user input, the proposed decision-making model predicts whether the task is computationally intensive or not. If the predicted result is computationally intensive, then based on the network parameters the proposed decision-making model predicts whether to offload or locally execute the task. According to the predicted result the task is either locally executed or offloaded to the edge server. The proposed method is implemented in a real-time environment, and the experimental results show that the proposed method has achieved above 90% prediction accuracy in offloading decision-making. The experimental results also present that the proposed offloading method reduces the response time and energy consumption of the user device by ~11-31% for computationally intensive tasks. A partial computation offloading method for federated learning is also proposed and implemented in this paper, where the devices which are unable to analyse the huge number of data samples, offload a part of their local datasets to the edge server. For secure data transmission, cryptography is used. The experimental results present that using encryption and decryption the total time is increased by only 0.05-0.16%. The results also present that the proposed partial computation offloading method for federated learning has achieved a prediction accuracy of above 98% for the global model.
Paper Structure (24 sections, 1 equation, 20 figures, 6 tables, 3 algorithms)

This paper contains 24 sections, 1 equation, 20 figures, 6 tables, 3 algorithms.

Table of Contents

  1. Introduction
  2. Motivation and Contributions
  3. Related Work
  4. Proposed Offloading Methods
  5. Proposed Offloading Decision-making Method
  6. Problem statement
  7. Proposed framework
  8. Computational complexity
  9. Proposed Federated Offloading Method
  10. Problem statement
  11. Proposed framework
  12. Computational complexity
  13. Total time consumption
  14. Performance Evaluation
  15. Performance of FLDec
  16. ...and 9 more sections

Figures (20)

  • Figure 1: System model of the proposed offloading decision-making process using FL
  • Figure 2: The proposed federated offloading method
  • Figure 3: Performance of the local and global models for predicting the task is computational intensive or not
  • Figure 4: Global model loss while predicting the task is computational intensive or not
  • Figure 5: Performance of the local and global models for predicting whether to offload or locally execute the task
  • ...and 15 more figures