Abstract:Integrated sensing and communication (ISAC) is one of the usage scenarios for the sixth generation (6G) wireless networks. In this paper, we study cooperative ISAC in cell-free multiple-input multiple-output (MIMO) systems, where multiple MIMO access points (APs) collaboratively provide communication services and perform multi-static sensing. We formulate an optimization problem for the ISAC beamforming design, which maximizes the achievable sum-rate while guaranteeing the sensing signal-to-noise ratio (SNR) requirement and total power constraint. Learning-based techniques are regarded as a promising approach for addressing such a nonconvex optimization problem. By taking the topology of cell-free MIMO systems into consideration, we propose a heterogeneous graph neural network (GNN), namely SACGNN, for ISAC beamforming design. The proposed SACGNN framework models the cell-free MIMO system for cooperative ISAC as a heterogeneous graph and employs a transformer-based heterogeneous message passing scheme to capture the important information of sensing and communication channels and propagate the information through the graph network. Simulation results demonstrate the performance gain of the proposed SACGNN framework over a conventional null-space projection based scheme and a deep neural network (DNN)-based baseline scheme.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is how to design a collaborative Integrated Sensing and Communication (ISAC) beamforming scheme in a Cell - Free Multiple - Input Multiple - Output (MIMO) system. Specifically, the goals of the paper are: 1. **Maximize the achievable sum - rate**: Ensure that the total transmission rate of the communication system is as high as possible under specific conditions. 2. **Guarantee the sensing Signal - to - Noise Ratio (SNR) requirement**: Ensure that the sensing signal used for target detection has a sufficient SNR to meet the requirements of sensing performance. 3. **Meet the total power constraint**: Ensure that the total transmit power of all Transmit - Access Points (APs) does not exceed a given maximum value. To achieve these goals, the authors propose a Heterogeneous Graph Neural Network (GNN), namely SACGNN (Sensing and Communication GNN), for ISAC beamforming design. SACGNN models the Cell - Free MIMO system as a heterogeneous graph and uses a Transformer - based heterogeneous message - passing scheme to capture important information of sensing and communication channels and propagate this information in the network. ### Problem Formulation The optimization problem can be formalized as: \[ \begin{aligned} & \text{maximize} & & R \\ & \text{subject to} & & \gamma^{(s)} \geq \gamma_{\min}, \\ & & & \|F_i\|_F^2 \leq P_{\max}, \quad i = 1, \ldots, N_T, \end{aligned} \] where: - \( R \) is the achievable sum - rate of all users; - \( \gamma^{(s)} \) is the sensing SNR; - \( \gamma_{\min} \) is the minimum sensing SNR requirement; - \( F_i \) is the beamforming matrix of the \( i \) - th transmit AP; - \( P_{\max} \) is the maximum transmit power of each transmit AP. ### Main Contributions 1. **Propose the SACGNN framework**: This framework models the Cell - Free MIMO system as a heterogeneous graph and uses a Transformer - based message - passing mechanism to handle complex communication and sensing tasks. 2. **Performance evaluation**: Through simulation, it is verified that compared with the traditional Null - Space Regularized Zero - Forcing (NS - RZF) method and the Deep Neural Network (DNN) - based method, the SACGNN framework can significantly improve the total rate of the system while meeting the sensing SNR requirements. In conclusion, this paper aims to solve the beamforming design problem for efficient communication and target detection simultaneously in a Cell - Free MIMO system, proposes the innovative SACGNN framework, and verifies its effectiveness and superiority through experiments.

Heterogeneous Graph Neural Network for Cooperative ISAC Beamforming in Cell-Free MIMO Systems

Learning Beamforming in Cell-Free Massive MIMO ISAC Systems

Integrated Sensing and Communication Enabled Multi-Device Multi-Target Cooperative Sensing: A Fairness-aware Design

Cooperative Beamforming and RISs Association for Multi-RISs Aided Multi-Users MmWave MIMO Systems through Graph Neural Networks

Sub-6GHz Assisted mmWave Hybrid Beamforming with Heterogeneous Graph Neural Network

Access Point Selection and Beamforming Design for Cell-Free Network: From Fractional Programming to GNN

A GNN Approach for Cell-Free Massive MIMO

Heterogeneous graph neural network for power allocation in multicarrier-division duplex cell-free massive MIMO systems

Joint User Pairing and Beamforming Design for NOMA-Aided CFMM-ISAC Systems

Cooperative Cell-Free ISAC Networks: Joint BS Mode Selection and Beamforming Design

A Homogeneous Graph Neural Network for Precoding and Power Allocation in Scalable Wireless Networks

Distributed Auto-Learning GNN for Multi-Cell Cluster-Free NOMA Communications

GNN-Aided User Association and Beam Selection for Mmwave-Integrated Heterogeneous Networks

GNN-Based Beamforming for Sum-Rate Maximization in MU-MISO Networks

Implementing Graph Neural Networks over Wireless Networks Via Over-the-Air Computing: A Joint Communication and Computation Framework

Cell-free ISAC Massive MIMO Systems with Capacity-Constrained Fronthaul Links

Distributed Combinatorial Optimization of Downlink User Assignment in mmWave Cell-free Massive MIMO Using Graph Neural Networks

GSM: A GNN-based Space-MIMO Framework for Direct-to-Cell Communications

Graph Neural Network based Active and Passive Beamforming for Distributed STAR-RIS-Assisted Multi-User MISO Systems

Distributed Graph Neural Network Design for Sum Ergodic Spectral Efficiency Maximization in Cell-Free Massive MIMO

Model-Based GNN Enabled Energy-Efficient Beamforming for Ultra-Dense Wireless Networks