Joint Optimization of Pilot Length, Pilot Assignment, and Power Allocation for Cell-free MIMO Systems with Graph Neural Networks

TL;DR

This paper introduces a deep learning framework using size-generalizable graph neural networks to jointly optimize pilot length, assignment, and power in cell-free MIMO systems, significantly reducing pilot contamination and improving spectral efficiency.

Contribution

It proposes a novel dual-timescale GNN framework with feature enhancement and contamination-aware attention for joint optimization, addressing the challenge of variable pilot length and unknown assignment size.

Findings

GNNs outperform existing methods in net-SE and inference speed.

The approach generalizes well across different problem scales and channels.

Significant reduction in pilot contamination and improved spectral efficiency.

Abstract

In user-centric cell-free multi-antenna systems, pilot contamination degrades spectral efficiency (SE) severely. To mitigate pilot contamination, existing works jointly optimize pilot assignment and power allocation by assuming fixed pilot length, which fail to balance pilot overhead against the contamination. To maximize net-SE, we jointly optimize pilot length, pilot assignment, and power allocation with deep learning. Since the pilot length is a variable, the size of pilot assignment matrix is unknown during the optimization. To cope with the challenge, we design size-generalizable graph neural networks (GNNs). We prove that pilot assignment policy is a one-to-many mapping, and improperly designed GNNs cannot learn the optimal policy. We tackle this issue by introducing feature enhancement. To improve learning performance, we design a contamination-aware attention mechanism for the GNNs. Given that pilot assignment and power allocation respectively depend on large- and small-scale channels, we develop a dual-timescale GNN framework to explore the potential. To reduce inference time, a single-timescale GNN is also designed. Simulation results show that the designed GNNs outperform existing methods in terms of net-SE, training complexity, and inference time, and can be well generalized across problem scales and channels.