Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Human-Centric Decision-Making in Cell-Less 6G Networks

Emma Chiaramello, Carla Fabiana Chiasserini, Francesco Malandrino, Alessandro Nordio, Marta Parazzini, Alvaro Valcarce

TL;DR

This work tackles human-centric decision-making in future cell-less networks by formulating a joint PoA-user association and PoA management problem that minimizes energy while guaranteeing user data rates and limiting EMF exposure. It introduces Cluster-then-Match (CtM), a three-step heuristic that clusters users, matches clusters to beams via a Hungarian algorithm, and then optimizes beam width and transmission power, all within a polynomial-time framework. Using ETSI TR138901-based channel models and anatomically varied exposure assessments, CtM demonstrates energy reductions of over 80% compared to MaxRate and maintains SAR well below ICNIRP limits across indoor and outdoor reference scenarios. Against ML-based benchmarks, CtM remains competitive, with DtM offering improvements given substantial training, while ML-only approaches underperform, highlighting the value of combining domain knowledge with learning for practical 6G deployment.

Abstract

In next-generation networks, cells will be replaced by a collection of points-of-access (PoAs), with overlapping coverage areas and/or different technologies. Along with a promise for greater performance and flexibility, this creates further pressure on network management algorithms, which must make joint decisions on (i) PoA-to-user association and (ii) PoA management. We solve this challenging problem through an efficient and effective solution concept called Cluster-then-Match (CtM). Importantly, CtM makes human-centric decisions, where pure network performance is balanced against metrics like energy consumption and electromagnetic field exposure, which concern all humans in the network area -- including those who are not network users. Through our performance evaluation, which leverages detailed models for EMF exposure estimation and standard-specified signal propagation models, we show that CtM outperforms state-of-the-art network management schemes, including those utilizing machine learning, reducing energy consumption by over 80%.

Human-Centric Decision-Making in Cell-Less 6G Networks

TL;DR

This work tackles human-centric decision-making in future cell-less networks by formulating a joint PoA-user association and PoA management problem that minimizes energy while guaranteeing user data rates and limiting EMF exposure. It introduces Cluster-then-Match (CtM), a three-step heuristic that clusters users, matches clusters to beams via a Hungarian algorithm, and then optimizes beam width and transmission power, all within a polynomial-time framework. Using ETSI TR138901-based channel models and anatomically varied exposure assessments, CtM demonstrates energy reductions of over 80% compared to MaxRate and maintains SAR well below ICNIRP limits across indoor and outdoor reference scenarios. Against ML-based benchmarks, CtM remains competitive, with DtM offering improvements given substantial training, while ML-only approaches underperform, highlighting the value of combining domain knowledge with learning for practical 6G deployment.

Abstract

In next-generation networks, cells will be replaced by a collection of points-of-access (PoAs), with overlapping coverage areas and/or different technologies. Along with a promise for greater performance and flexibility, this creates further pressure on network management algorithms, which must make joint decisions on (i) PoA-to-user association and (ii) PoA management. We solve this challenging problem through an efficient and effective solution concept called Cluster-then-Match (CtM). Importantly, CtM makes human-centric decisions, where pure network performance is balanced against metrics like energy consumption and electromagnetic field exposure, which concern all humans in the network area -- including those who are not network users. Through our performance evaluation, which leverages detailed models for EMF exposure estimation and standard-specified signal propagation models, we show that CtM outperforms state-of-the-art network management schemes, including those utilizing machine learning, reducing energy consumption by over 80%.
Paper Structure (20 sections, 8 equations, 11 figures, 1 table)

This paper contains 20 sections, 8 equations, 11 figures, 1 table.

Table of Contents

  1. Introduction
  2. System Model
  3. Optimization Formulation
  4. Objective and requirements.
  5. Problem analysis
  6. The CtM Strategy
  7. Algorithm analysis
  8. Reference Scenario
  9. Antenna model and beamforming
  10. Channel impulse response
  11. SINR, rate, and power density at the target
  12. Exposure model
  13. Reference topologies
  14. Performance Evaluation
  15. MaxRate and ML-based benchmarks
  16. ...and 5 more sections

Figures (11)

  • Figure 1: A simple indoor cell-less network. Two points-of-access (PoAs) serve their end users through one or more beams; one of the users can be served by two PoAs. A human is irradiated by the beam from PoA #1 to User #2; humans will nonetheless incur EMF exposure whether or not they are network users.
  • Figure 2: The three main steps of the CtM strategy: clustering of the end users, cluster-to-PoA assignment, optimizing beam width and transmission power levels.
  • Figure 3: Global coordinate system (GCS) (left), hall model for the InF-DH indoor scenario with colored objects representing clutters (center), and the detailed anatomical models used for electromagnetic field exposure assessment (right).
  • Figure 4: The ML-based benchmarks we compare against: DtM (top), replacing stage 1 of CtM with DNN-based clustering ohi2020autoembedder, and ML-only (bottom), using a DNN to make all decisions. Additions and changes with respect to CtM (Fig. \ref{['fig:ctm']}) are marked in purple.
  • Figure 5: Scheme of the simulator used for performance evaluation.
  • ...and 6 more figures