Analysis of Terahertz (THz) Frequency Propagation and Link Design for Federated Learning in 6G Wireless Systems | IEEE Journals & Magazine | IEEE Xplore

Analysis of Terahertz (THz) Frequency Propagation and Link Design for Federated Learning in 6G Wireless Systems


To address the resource issue in Federated Learning, we propose the use of Terahertz higher bandwidth spectrum for Wireless Backhaul. In this article, we discuss the prop...

Abstract:

Increased throughput demands in emerging services drive a rapid shift from 5G to 6G, posing interdisciplinary challenges in wireless communication stacks. This impacts ne...Show More

Abstract:

Increased throughput demands in emerging services drive a rapid shift from 5G to 6G, posing interdisciplinary challenges in wireless communication stacks. This impacts network modeling and deployment, with AI playing a crucial role. Terahertz (THz) communication spectrum and Federated Learning (FL) gain traction in the 6G paradigm. FL, a decentralized approach, emphasizes data confidentiality and security in wireless networks. The THz spectrum (0.1 to 10 THz) is vital for ultra-broadband wireless systems beyond 5G, offering high data rates. THz waves hold promise for short-distance broadband wireless access, acting as optical network bridges in challenging environments. Despite limited range and penetration, THz technology maximizes spectrum usage, enhancing transmission security. This article offers a concise overview of the Terahertz (THz) spectrum in fixed wireless communication, examining applications and future possibilities. It conducts a thorough analysis, comparing THz with microwave and mm-wave spectra regarding various factors. THz can significantly improve data rates, up to 10 times, reaching 100 Gbps. Spreading loss is around 150 dB within 1 km, doubling to over 300 dB at 2 km. For 300 GHz, it provides a Receive Signal Level (RSL) of −43.57 dBm; increasing path length results in a straight decrease to −56 dB for RSL. These highlights lead to the conclusion that a Terahertz-based network has the potential to enhance convergence time and reduce training loss in Federated Learning, particularly in 1 km links, due to favorable conditions for efficient data transmission. We propose leveraging the largely untapped THz frequency band to enhance FL communication. In the healthcare sector, we introduce FL, relying on a wireless backhaul infrastructure and THz-based wireless backhaul with a Virtual Private Network (VPN). Hospitals are identified as the designated end-users who employ a private network through service providers’ wireless backhaul network to enhance...
To address the resource issue in Federated Learning, we propose the use of Terahertz higher bandwidth spectrum for Wireless Backhaul. In this article, we discuss the prop...
Published in: IEEE Access ( Volume: 12)
Page(s): 23782 - 23797
Date of Publication: 06 February 2024
Electronic ISSN: 2169-3536

Funding Agency:


References

References is not available for this document.