Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks

Jungmin Kwon; Hyunggon Park

doi:10.1109/ICUFN65838.2025.11169983

Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks

Jungmin Kwon
, Hyunggon Park

Electronic and Electrical Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

In wireless federated learning (FL) systems, the use of the User Datagram Protocol (UDP) has been explored to reduce communication overhead by avoiding retransmissions. However, UDP-based transmission is inherently unreliable and can lead to packet loss, causing parts of the model parameters to be lost and thereby degrading the overall training performance. To address this issue, network coding (NC) techniques have been proposed as a complementary solution that linearly combines packets to improve both reliability and efficiency. In this paper, we incorporate several NC algorithms into a centralized FL system and experimentally evaluate how they affect model accuracy and communication efficiency under unreliable communication conditions.

Original language	English
Title of host publication	ICUFN 2025 - 16th International Conference on Ubiquitous and Future Networks
Publisher	IEEE Computer Society
Pages	495-497
Number of pages	3
ISBN (Electronic)	9798331524876
DOIs	https://doi.org/10.1109/ICUFN65838.2025.11169983
State	Published - 2025
Event	16th International Conference on Ubiquitous and Future Networks, ICUFN 2025 - Hybrid, Lisbon, Portugal Duration: 8 Jul 2025 → 11 Jul 2025

Publication series

Name	International Conference on Ubiquitous and Future Networks, ICUFN
ISSN (Print)	2165-8528
ISSN (Electronic)	2165-8536

Conference

Conference	16th International Conference on Ubiquitous and Future Networks, ICUFN 2025
Country/Territory	Portugal
City	Hybrid, Lisbon
Period	8/07/25 → 11/07/25

Bibliographical note

Publisher Copyright:
© 2025 IEEE.

Keywords

Federated learning
network coding
unreliable network
user datagram protocol

Access to Document

10.1109/ICUFN65838.2025.11169983

Cite this

Kwon, J., & Park, H. (2025). Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks. In ICUFN 2025 - 16th International Conference on Ubiquitous and Future Networks (pp. 495-497). (International Conference on Ubiquitous and Future Networks, ICUFN). IEEE Computer Society. https://doi.org/10.1109/ICUFN65838.2025.11169983

@inproceedings{fb25925f2de44e2c904812595092d83b,

title = "Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks",

abstract = "In wireless federated learning (FL) systems, the use of the User Datagram Protocol (UDP) has been explored to reduce communication overhead by avoiding retransmissions. However, UDP-based transmission is inherently unreliable and can lead to packet loss, causing parts of the model parameters to be lost and thereby degrading the overall training performance. To address this issue, network coding (NC) techniques have been proposed as a complementary solution that linearly combines packets to improve both reliability and efficiency. In this paper, we incorporate several NC algorithms into a centralized FL system and experimentally evaluate how they affect model accuracy and communication efficiency under unreliable communication conditions.",

keywords = "Federated learning, network coding, unreliable network, user datagram protocol",

author = "Jungmin Kwon and Hyunggon Park",

note = "Publisher Copyright: {\textcopyright} 2025 IEEE.; 16th International Conference on Ubiquitous and Future Networks, ICUFN 2025 ; Conference date: 08-07-2025 Through 11-07-2025",

year = "2025",

doi = "10.1109/ICUFN65838.2025.11169983",

language = "English",

series = "International Conference on Ubiquitous and Future Networks, ICUFN",

publisher = "IEEE Computer Society",

pages = "495--497",

booktitle = "ICUFN 2025 - 16th International Conference on Ubiquitous and Future Networks",

}

Kwon, J & Park, H 2025, Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks. in ICUFN 2025 - 16th International Conference on Ubiquitous and Future Networks. International Conference on Ubiquitous and Future Networks, ICUFN, IEEE Computer Society, pp. 495-497, 16th International Conference on Ubiquitous and Future Networks, ICUFN 2025, Hybrid, Lisbon, Portugal, 8/07/25. https://doi.org/10.1109/ICUFN65838.2025.11169983

Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks. / Kwon, Jungmin; Park, Hyunggon.
ICUFN 2025 - 16th International Conference on Ubiquitous and Future Networks. IEEE Computer Society, 2025. p. 495-497 (International Conference on Ubiquitous and Future Networks, ICUFN).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks

AU - Kwon, Jungmin

AU - Park, Hyunggon

PY - 2025

Y1 - 2025

N2 - In wireless federated learning (FL) systems, the use of the User Datagram Protocol (UDP) has been explored to reduce communication overhead by avoiding retransmissions. However, UDP-based transmission is inherently unreliable and can lead to packet loss, causing parts of the model parameters to be lost and thereby degrading the overall training performance. To address this issue, network coding (NC) techniques have been proposed as a complementary solution that linearly combines packets to improve both reliability and efficiency. In this paper, we incorporate several NC algorithms into a centralized FL system and experimentally evaluate how they affect model accuracy and communication efficiency under unreliable communication conditions.

AB - In wireless federated learning (FL) systems, the use of the User Datagram Protocol (UDP) has been explored to reduce communication overhead by avoiding retransmissions. However, UDP-based transmission is inherently unreliable and can lead to packet loss, causing parts of the model parameters to be lost and thereby degrading the overall training performance. To address this issue, network coding (NC) techniques have been proposed as a complementary solution that linearly combines packets to improve both reliability and efficiency. In this paper, we incorporate several NC algorithms into a centralized FL system and experimentally evaluate how they affect model accuracy and communication efficiency under unreliable communication conditions.

KW - Federated learning

KW - network coding

KW - unreliable network

KW - user datagram protocol

UR - https://www.scopus.com/pages/publications/105018743791

U2 - 10.1109/ICUFN65838.2025.11169983

DO - 10.1109/ICUFN65838.2025.11169983

M3 - Conference contribution

AN - SCOPUS:105018743791

T3 - International Conference on Ubiquitous and Future Networks, ICUFN

SP - 495

EP - 497

BT - ICUFN 2025 - 16th International Conference on Ubiquitous and Future Networks

PB - IEEE Computer Society

T2 - 16th International Conference on Ubiquitous and Future Networks, ICUFN 2025

Y2 - 8 July 2025 through 11 July 2025

ER -

Comparative Analysis of Network Coding Algorithms in Centralized Federated Learning Over Unreliable Networks

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this