Cell-Free Multi-User MIMO Equalization via In-Context Learning

TL;DR

The paper tackles uplink data symbol detection in cell-free MIMO networks under limited fronthaul capacity and pilot contamination. It proposes a decoder-only transformer that uses in-context learning (ICL) with task descriptors $f(\tau)$ and context $\mathcal{D}_{\tau,k}$ to estimate data symbols $\hat{x}_{k|\theta}$ without any parameter updates. The approach relies on carefully designed prompts that incorporate pilot sequences, large-scale fading information, and modulation, enabling rapid adaptation across diverse tasks. Pre-training over thousands of tasks shows that ICL can outperform centralized LMMSE, especially when fronthaul capacity is tight, making it practical for disaggregated RAN architectures.

Abstract

Large pre-trained sequence models, such as transformers, excel as few-shot learners capable of in-context learning (ICL). In ICL, a model is trained to adapt its operation to a new task based on limited contextual information, typically in the form of a few training examples for the given task. Previous work has explored the use of ICL for channel equalization in single-user multi-input and multiple-output (MIMO) systems. In this work, we demonstrate that ICL can be also used to tackle the problem of multi-user equalization in cell-free MIMO systems with limited fronthaul capacity. In this scenario, a task is defined by channel statistics, signal-to-noise ratio, and modulation schemes. The context encompasses the users' pilot sequences, the corresponding quantized received signals, and the current received data signal. Different prompt design strategies are proposed and evaluated that encompass also large-scale fading and modulation information. Experiments demonstrate that ICL-based equalization provides estimates with lower mean squared error as compared to the linear minimum mean squared error equalizer, especially in the presence of limited fronthaul capacity and pilot contamination.

Figures (4)

• Figure 1: ICL-based channel equalization for cell-free MIMO systems.
• Figure 2: Illustration of the ICL-equalizer based on decoder-only transformers.
• Figure 3: Test mean squared error as a function of the fronthaul capacity $b$ for an orthogonal pilot assignment on the left panel, and for a randomized assignment with pilot reuse on the right ($\text{SNR}=24$ dB).
• Figure 4: Test mean squared error as a function of the average SNR and for different levels of pilot reuse (Fronthaul capacity $b=8$ bits per symbol).