A Trust and Explainable Federated Deep Learning Framework in Zero Touch B5G Networks
Open accessThe DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
The emergent Zero touch Service and Management (ZSM) paradigm aims to automate the orchestration and management of running network slices, in Beyond 5G networks (B5G), with an unprecedented level of scalability. To achieve this vision, ZSM calls for a large usage of advanced deep learning algorithms, in order to dynamically build efficient decisions. In this context, Federated deep Learning (FL) proved their efficiency in not only building collaborative deep learning models, among several network slices, but also ensuring the privacy and isolation of such network slices. Indeed, FL-based solutions give “machine-centric” decisions about running network slices and their performance, which will be then executed/applied by managers, i.e., slice manager staff/module. However, FL-enabled solutions do not provide any details about why and how such decisions were made, and thus such decisions cannot be properly trusted/understood by slice managers. To alleviate this issue, we leverage eXplainable Artificial Intelligence (XAI) paradigm that aims to improve the transparency of black-box FL decision-making process. In particular, XAI helps to explain the FL-based decisions to make them interpretable/trustable by network slices managers. In this paper, we design a novel XAI-powered framework to explain FL-based decisions. We first build a deep learning model in federated way, to predict key performance indicators (KPI) of network slices. Our FL-based KPI prediction is useful for the configuration and the management of network slice lifecycle, especially for the Service Level Agreement (SLA) violation and the network slice re-configuration. Then, we develop several XAI models on the top of our FL-based model, such as SHapley Additive exPlanations (SHAP), Local Interpretable Model-agnostic Explanations (LIME), RuleFit, and Partial Dependence Plot (PDP), to enhance the level of trust, credibility (of the local data/model), transparency, and explanation of the FL-based decisions, while adhering the data privacy, to different B5G network stakeholders, such as slice managers. Experiments results show the efficiency of our XAI-powered framework, to explain FL-based decisions related to latency KPI predictions.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“It combines previously proposed explainability methods, such as LIME [22] and Deep Learning Important FeaTures (DeepLIFT) [41], to approximate the Shapley values.”
From Explainable AI-Aided Feature Selection and Model Reduction for DRL-Based V2X Resource Allocation (2025) · cites this paper as [22]
In the citing paper
“XAI-based feature selection methods are generally architecture-specific, data-dependent, and provide local explanations [22].”
From Explainable AI-Aided Feature Selection and Model Reduction for DRL-Based V2X Resource Allocation (2025) · cites this paper as [22]
In the citing paper
“Second, SHAP combines the local interpretable model-agnostic explanations (LIME) [22] method, which explains individual predictions using an interpretable model, with Shapley values. This integration allows for faster computation of explanations for complex learning models than directly calculating Shapley values.”
From Explainable AI-Aided Feature Selection and Model Reduction for DRL-Based V2X Resource Allocation (2025) · cites this paper as [22]
Citations
Cites
Interpret Federated Learning with Shapley ValuesPreprint
Wang, G. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1905.04519 · Cited by 117
Secure Federated Learning in 5G Mobile Networks
Isaksson, M., & Norrman, K. (2020) · Globecom · doi:10.1109/globecom42002.2020.9322479 · Cited by 39
An end‐to‐end trusted architecture for network slicing in 5G and beyond networks
Saad, S. B., Ksentini, A., & Brik, B. (2021) · Security and Privacy · doi:10.1002/spy2.186 · Cited by 33
Self-citationA Trust architecture for the SLA management in 5G networks
Saad, S. B., Ksentini, A., & Brik, B. (2021) · International Conference on Communications · doi:10.1109/icc42927.2021.9500990 · Cited by 18
Self-citationExplainable Federated Learning for Taxi Travel Time Prediction
Fiošina, J. (2021) · International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS) · doi:10.5220/0010485600002932 · Cited by 12
Research works printed but not linked: 1
- Molnar, 2022, Interpretable Machine Learning A Guide for Making Black Box Models Explainable
Cited by
Advancing 6G: Survey for Explainable AI on Communications and Network Slicing
Sun, H., Liu, Y., Al-Tahmeesschi, A., Nag, A., Soleimanpour-Moghadam, M., Canberk, B., Arslan, H., & Ahmadi, H. (2025) · IEEE Open Journal of the Communications Society · doi:10.1109/ojcoms.2025.3534626 · Cited by 151
Split Federated Learning for 6G Enabled-Networks: Requirements, Challenges, and Future Directions
Hafi, H., Brik, B., Frangoudis, P. A., Ksentini, A., & Bagaa, M. (2024) · IEEE Access · doi:10.1109/access.2024.3351600 · Cited by 90
“In other words, XAI helps to explain the SFL-based decisions to make them trustable and interpretable by the different 6G actors [223].”
Self-citationBrik, B., Chergui, H., Zanzi, L., Devoti, F., Ksentini, A., Siddiqui, M. S., Costa‐Pérez, X., & Verikoukis, C. V. (2024) · IEEE Communications Surveys & Tutorials · doi:10.1109/comst.2024.3510543 · Cited by 142
Self-citationFederated Learning of XAI Models in Healthcare: A Case Study on Parkinson’s Disease
Ducange, P., Marcelloni, F., Renda, A., & Ruffini, F. (2024) · Cognitive Computation · doi:10.1007/s12559-024-10332-x · Cited by 54
Explanation-Guided Deep Reinforcement Learning for Trustworthy 6G RAN Slicing
Rezazadeh, F., Chergui, H., & Mangues‐Bafalluy, J. (2023) · IEEE International Conference on Communications workshops · doi:10.1109/iccworkshops57953.2023.10283684 · Cited by 47
Metadata
- Authors
- Ben Saad, S., Brik, B., & Ksentini, A.
- Affiliations
- Eurecom,Sophia Antipolis,France
- Université de Bourgogne (FR)
- EURECOM (FR)
- Venue
- GLOBECOM 2022 - 2022 IEEE Global Communications Conference
- Publisher
- IEEE
- Published
- 2022-12-04
- Pages
- 1037-1042
- ISSN
- 1930-529X (ISSN-L)
- OpenAlex
- W4315605901
- Cited by
- 21
- References
- 5
- Citations by year
- 2026: 3 · 2025: 3 · 2024: 9 · 2023: 6
- FWCI
- 2.49
- Type
- conference paper
- Language
- en
- Open access
- green
- hal.science
- Locations
- Globecom (publishedVersion)
- HAL (submittedVersion, other-oa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Computer Networks and Communications › Software-Defined Networks and 5G
- Physical Sciences › Computer Science › Artificial Intelligence › Privacy-Preserving Technologies in Data
- Keywords
- federated deep learning, explainable AI, beyond 5G, network slice management, SLA violation, SHAP, lime, RuleFit, partial dependence plots, latency prediction, data privacy
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: A Trust and Explainable Federated Deep Learning Framework in Zero Touch B5G Networks. https://app.cochat.ai/papers/10.1109/globecom48099.2022.10001371
- The paper
- Ben Saad, S., Brik, B., & Ksentini, A. (2022). A Trust and Explainable Federated Deep Learning Framework in Zero Touch B5G Networks. GLOBECOM 2022 - 2022 IEEE Global Communications Conference. https://doi.org/10.1109/globecom48099.2022.10001371
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.