Resilient Cell-Free MIMO ISAC Under Uncertainties: An Impairment-Aware Graph Neural Network Framework

Oways Alsoloh; Robert-Jeron Reifert; Aymen Khaleel; Stefan Roth; Aydin Sezgin

eess.SPNov 2, 2026

Resilient Cell-Free MIMO ISAC Under Uncertainties: An Impairment-Aware Graph Neural Network Framework

Authors: Oways Alsoloh, Robert-Jeron Reifert, Aymen Khaleel, Stefan Roth, Aydin Sezgin

Organizations: Institute of Digital Communication Systems, Ruhr University Bochum, Bochum, Germany

Abstract

Beamforming designs for integrated sensing and communication (ISAC) in cell-free multiple-input multiple-output (MIMO) systems face the challenge to stay reliable under uncertainties from both channel state information (CSI) as well as clutter. To this end, this paper proposes an impairment-aware heterogeneous graph neural network (GNN) framework for beamforming that obtains the CSI error level and the clutter covariance as edge features, which enables a single trained model to adapt to both impairments. In particular, the GNN is trained on a low quantile of the minimum user rate with a clutter-aware signal-to-clutter-plus-noise ratio, balanced by a normalized Tchebycheff scalarization. Simulation results show that the proposed GNN framework outperforms the considered baselines in sensing-only and communication-only settings. Depending on the operating point, a single GNN model attains approximately 97% of each baseline's rate while serving both functionalities at once. The model further generalizes to unseen CSI error levels and clutter counts and renders the ISAC system resilient.

Figures & tables

Panel of Figure 1 — Fig. 1: Cell-free MIMO ISAC scenario, with distributed APs serving the UEs and sensing a target in the presence of clutter.

Panel of Figure 2 — Fig. 2: Proposed impairment-aware GNN framework, which maps the clutter-aware and CSI error-aware edge features to the beamformers $\{\mathbf{w}_{mu}\}$ , trained with the Tchebycheff loss of ( 18 ) and ( 19 ).

System parameters		GNN and optimization parameters
APs	$M=3$	Layers	$L=4$
Antennas per ULA	$N=4$	Hidden size	$256$
Users	$K=12$	Mini-batch size	$B=100$
Sensing streams	$S=\|\mathcal{S}\|=1$	Learning rate	$2\times 10^{-4}$
Total Tx power	$P_{\mathrm{tot}}=6$	Training samples	$20000$
Per-AP power	$P_{m}=2$	Val. samples	$2000$

TABLE I: Main simulation and training parameters.

Panel of Figure 4 — Fig. 3: Communication–sensing operating plane.

Panel of Figure 5 — Fig. 4: Robust communication and sensing performance (a) under CSI uncertainty and (b) versus the number of clutter reflectors.

Integrated Sensing and Communication

Integrated sensing and communication combines wireless data links with radar-like perception using shared signals, spectrum, and hardware.

See all 54 papers

Beamforming Design Via GNN in mmWave Cell-Free Massive MIMO Using Sub-6 GHz CSI
Aug 31, 2026Sina Tavakolian, Abolfazl Zakeri, Ahmed Alkhateeb +2Graph Neural Networks Wireless Communications
⋆Centre for Wireless Communications, University of Oulu, P.O.Box 4500, FI-90014, Finland · †School of Electrical, Computer, and Energy Engineering, Arizona State University, AZ, USA
Feasible but Not Safe: Constraint Violations and Report-Channel Attacks in Learned Cell-Free ISAC Association
Sep 2, 2026Mehdi Zafari, Iman Mohammadi, A. Lee SwindlehurstGraph Neural Networks Integrated Sensing and Communication
University of California Irvine · Irvine, CA, USA
Graph Neural Networks for Resource Allocation in Integrated Sensing and Communication
Nov 5, 2026Junyu Liao, Lili Chen, Tian Han +1Vehicular Networks Wireless Resource Allocation
Department of Electrical and Electronic Engineering, University of Melbourne, Australia · Department of Electrical and Computer Systems Engineering, Monash University, Australia

Aug 31, 2026J/K move · Enter open · S save

Beamforming Design Via GNN in mmWave Cell-Free Massive MIMO Using Sub-6 GHz CSI

Sina Tavakolian, Abolfazl Zakeri, Ahmed Alkhateeb, Markku Juntti+1

⋆Centre for Wireless Communications, University of Oulu, P.O.Box 4500, FI-90014, Finland · †School of Electrical, Computer, and Energy Engineering, Arizona State University, AZ, USA

Graph Neural Networks Wireless Communications Millimeter-Wave Communications

Open paper AskPDF arXiv

Beamforming methods in millimeter-wave (mmWave) cell-free massive multiple-input multiple-output (CFmMIMO) systems require accurate channel state information (CSI), whose acquisition entails significant training overhead. This paper shows that fully digital cell-free mmWave beamforming can be effectively learned from sub-6 GHz CSI using a graph neural network (GNN). Specifically, we represent a CFmMIMO system as a wireless graph, and the GNN is trained to approximate beamformers that maximize the downlink sum-rate based on the available sub-6 GHz CSI. A message-passing mechanism is proposed to capture inter-user interference and inter-base-station cooperation across different network topologies. Simulation results demonstrate that the proposed sub-6 GHz-assisted GNN-based beamformer achieves competitive and often superior sum-rate performance compared to classical baselines that rely on full mmWave CSI.

Abstract

Integrated Sensing and Communication

Explore similar work

Beamforming Design Via GNN in mmWave Cell-Free Massive MIMO Using Sub-6 GHz CSI

Feasible but Not Safe: Constraint Violations and Report-Channel Attacks in Learned Cell-Free ISAC Association

Graph Neural Networks for Resource Allocation in Integrated Sensing and Communication