Deep Learning for Uplink CSI-based Downlink Precoding in FDD massive MIMO Evaluated on Indoor Measurements

TL;DR

This paper explores neural network-based methods to predict downlink CSI from uplink CSI in FDD massive MIMO systems, validated on real indoor measurements, addressing a key challenge in CSI acquisition.

Contribution

It introduces neural network approaches for UL to DL CSI mapping, along with a novel evaluation scheme for generalization in real-world indoor environments.

Findings

Neural networks outperform classical methods in CSI prediction.

The evaluation scheme effectively distinguishes known and unseen environments.

Results demonstrate feasibility of DL CSI prediction in static indoor settings.

Abstract

When operating massive multiple-input multiple-output (MIMO) systems with uplink (UL) and downlink (DL) channels at different frequencies (frequency division duplex (FDD) operation), acquisition of channel state information (CSI) for downlink precoding is a major challenge. Since, barring transceiver impairments, both UL and DL CSI are determined by the physical environment surrounding transmitter and receiver, it stands to reason that, for a static environment, a mapping from UL CSI to DL CSI may exist. First, we propose to use various neural network (NN)-based approaches that learn this mapping and provide baselines using classical signal processing. Second, we introduce a scheme to evaluate the performance and quality of generalization of all approaches, distinguishing between known and previously unseen physical locations. Third, we evaluate all approaches on a real-world indoor dataset collected with a 32-antenna channel sounder.