Graph neural networks reduce data load for distributed MIMO wireless systems
GNN-Based Global CSI Reconstruction for Fronthaul-Limited Distributed MIMO Systems
Information Theory
Summary
Wireless systems with many antennas need lots of data to understand the communication channels between devices. Sending all this data back to a central computer can overload the system. The authors designed a smart method that selects only some antenna data to send, then uses a graph neural network to fill in the missing parts. This approach keeps the system accurate while lowering the data needed for transmission.
What this means in practice
- •For wireless network engineers: Improve message passing in distributed antenna systems by reconstructing full channel data from partial antenna uploads to reduce network load.
- •For cellular infrastructure operators: Lower fronthaul bandwidth requirements for distributed MIMO setups by selectively uploading antenna data and reconstructing channels centrally.
Authors
Haojin Li, Kaiqian Qu, Anbang Zhang, Chen Sun, Wenqi Zhang, Haijun Zhang
Abstract
Global channel state information (CSI) acquisition is essential for cooperative precoding in distributed multiple-input multiple-output (DMIMO) systems, but uploading full instantaneous CSI from all distributed antennas creates heavy fronthaul overhead. This paper proposes a fronthaul-efficient acquisition framework based on graph neural network (GNN) reconstruction and task-driven antenna selection. Each transmission and reception point (TRP) uploads only selected antenna CSI, while the centralized unit (CU) reconstructs the full global CSI from partial observations. A universal mask-conditioned GNN is trained with random upload masks, used to evaluate antenna subsets under a fronthaul budget, and then fine-tuned for the selected deployment mask. Simulation results show improved CSI reconstruction accuracy with lower fronthaul and pilot overhead.