User Association in Coexisting RF and TeraHertz Networks in 6G

TL;DR

This paper introduces two novel user association algorithms for coexisting RF and THz networks in 6G, aiming to balance network load and improve data rates by considering frequency-specific heterogeneity.

Contribution

The paper proposes new user association algorithms that minimize traffic load standard deviation in RF and THz networks, accounting for their heterogeneity, unlike traditional clustering methods.

Findings

Algorithms outperform classical methods in data rate

Enhanced traffic load balancing achieved

Improved user fairness demonstrated

Abstract

While fifth generation (5G) networks are ready for deployment, discussions over sixth generation (6G) networks are down the road. Since high frequencies like terahertz (THz) will be central to 6G, in this paper, we propose two user association (UE) algorithms considering a coexisting RF and THz network that balances the traffic load across the network by minimizing the standard deviation of the network traffic load. Our algorithms capture the heterogeneity observed at RF and THz frequencies such as transmission bandwidth, molecular absorption, transmit powers, etc. Unlike typical unsupervised clustering algorithms (e.g. k-means, k-medoid, etc.) that search for appropriate cluster centers' locations, our algorithms identify the appropriate UEs to be associated to a certain BS such that the overall network load standard deviation (STD) can be minimized subject to users' rate constraints. In particular, our algorithms cluster UEs to every base station (BS) such that the traffic load across the network can be balanced, i.e., by minimizing the STD of network traffic load. Numerical results show that the proposed algorithms outperform the classical user association algorithms in terms of data rate, traffic load balancing, and user's fairness.