Toward Human-Perceptive 6G: Privacy-Preserving Network Intelligence from Human Dynamics

Abstract

Future mobile network management will depend heavily on the ability to learn and anticipate human behavior, including mobility and traffic demand, while preserving user privacy. Federated learning enables distributed spatio-temporal modeling and edge-based training without sharing raw data. However, existing approaches are mono-perspective, addressing mobility prediction, traffic forecasting, and network performance optimization as isolated tasks, with limited integration into network control and resource orchestration. This paper presents a vision architecture that integrates into the control loop of AI-native 6G networks, multi-perspective spatio-temporal representations of user mobility, traffic behavior, and network performance, derived through human-aware federated learning. The proposed architecture is an initial step toward scalable, privacy-preserving, and human-perceptive intelligence for 6G resource management.

Equitable Partition Realizability for Dynamics-preserving and Privacy-aware Network Reconstruction
Riccardo Porcedda
arXiv15 Sep 2026PreprintCC BY 4.0Full textStatisticsarXiv
Privacy Preserving Gossip Learning
Erkan Bayram, Mohamed-Ali Belabbas, Tamer Başar
arXiv13 Sep 2026PreprintarXiv licenceFull textOptimisation & controlarXiv
Privacy-Preserving Causal Meta-Mediation Analysis with Survival Outcomes
Marie-Félicia Beclin, Tat-Thang Vo
arXiv10 Sep 2026PreprintarXiv licenceFull textStatisticsarXiv
Design of Privacy-Preserving Estimators Under Prior Information Exposure
Le Liu, Yu Kawano, Ming Cao
arXiv12 Oct 2026PreprintarXiv licenceFull textGeneral mathematicsarXiv
SSTQ:Privacy-Preserving Vector Quantization via Subsampled Stochastic TurboQuant
Adel Javanmard, David P. Woodruff, Vahab Mirrokni
arXiv28 Sep 2026PreprintCC BY 4.0Full textStatisticsarXiv

Cite

BibTeX RIS

BibTeX

@misc{guevara2026toward,
  title = {{Toward Human-Perceptive 6G: Privacy-Preserving Network Intelligence from Human Dynamics}},
  author = {Judy C. Guevara and Anne Josiane Kouam and Nadjib Achir and Aline Carneiro Viana},
  howpublished = {HAL},
  year = {2026},
  month = jun,
  url = {https://inria.hal.science/hal-05742398v2},
  note = {CC BY-NC-ND 4.0},
}

RIS

TY  - UNPB
TI  - Toward Human-Perceptive 6G: Privacy-Preserving Network Intelligence from Human Dynamics
AU  - Judy C. Guevara
AU  - Anne Josiane Kouam
AU  - Nadjib Achir
AU  - Aline Carneiro Viana
PY  - 2026
DA  - 2026/06/05
PB  - HAL
UR  - https://inria.hal.science/hal-05742398v2
AB  - Future mobile network management will depend heavily on the ability to learn and anticipate human behavior, including mobility and traffic demand, while preserving user privacy. Federated learning enables distributed spatio-temporal modeling and edge-based training without sharing raw data. However, existing approaches are mono-perspective, addressing mobility prediction, traffic forecasting, and network performance optimization as isolated tasks, with limited integration into network control and resource orchestration. This paper presents a vision architecture that integrates into the control loop of AI-native 6G networks, multi-perspective spatio-temporal representations of user mobility, traffic behavior, and network performance, derived through human-aware federated learning. The proposed architecture is an initial step toward scalable, privacy-preserving, and human-perceptive intelligence for 6G resource management.
ER  -

Abstract

Related

Cite