TY - GEN
T1 - Federated Learning via Lattice Joint Source-Channel Coding
AU - Azimi-Abarghouyi, Seyed Mohammad
AU - Varshney, Lav R.
N1 - Publisher Copyright:
© 2024 IEEE.
PY - 2024
Y1 - 2024
N2 - This paper introduces a universal federated learning framework that enables over-the-air computation via digital communications, using a new joint source-channel coding scheme. Without relying on channel state information at devices, this scheme employs lattice codes to both quantize model parameters and exploit interference from the devices. A novel two-layer receiver structure at the server is designed to reliably decode an integer combination of the quantized model parameters as a lattice point for the purpose of aggregation. Numerical experiments validate the effectiveness of the proposed scheme. Even with the challenges posed by channel conditions and device heterogeneity, the proposed scheme markedly surpasses other over-the-air FL strategies.
AB - This paper introduces a universal federated learning framework that enables over-the-air computation via digital communications, using a new joint source-channel coding scheme. Without relying on channel state information at devices, this scheme employs lattice codes to both quantize model parameters and exploit interference from the devices. A novel two-layer receiver structure at the server is designed to reliably decode an integer combination of the quantized model parameters as a lattice point for the purpose of aggregation. Numerical experiments validate the effectiveness of the proposed scheme. Even with the challenges posed by channel conditions and device heterogeneity, the proposed scheme markedly surpasses other over-the-air FL strategies.
KW - Federated learning
KW - digital communications
KW - joint source-channel coding
KW - lattice codes
KW - over-the-air computation
UR - http://www.scopus.com/inward/record.url?scp=85202860688&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85202860688&partnerID=8YFLogxK
U2 - 10.1109/ISIT57864.2024.10619502
DO - 10.1109/ISIT57864.2024.10619502
M3 - Conference contribution
AN - SCOPUS:85202860688
T3 - IEEE International Symposium on Information Theory - Proceedings
SP - 1623
EP - 1628
BT - 2024 IEEE International Symposium on Information Theory, ISIT 2024 - Proceedings
PB - Institute of Electrical and Electronics Engineers Inc.
T2 - 2024 IEEE International Symposium on Information Theory, ISIT 2024
Y2 - 7 July 2024 through 12 July 2024
ER -