Variable-Length Wideband CSI Feedback via Loewner Interpolation and Deep Learning

TL;DR

The paper addresses the challenge of downlink CSI feedback in wideband FDD massive MIMO for the U6G band by introducing a variable-length framework that blends Loewner Interpolation in the frequency domain with a downstream rateless DL-based spatial compressor (LI-MORNet). MOR reduces the LI model to a compact descriptor set, while a cascaded, sparsity-enhanced spatial auto-encoder provides flexible, progressive reconstruction; a robust quantization scheme ensures resilience to quantization errors and tail erasures. The approach yields higher CSI reconstruction accuracy at equal or lower feedback overhead and improves spectral efficiency compared with DL and codebook-based baselines, with lower neural complexity and memory. The framework offers a practical, scalable solution for wideband CSI feedback in next-generation wireless systems, enabling adaptive feedback together with robust performance under quantization and erasures.

Abstract

In this paper, we propose a variable-length wideband channel state information (CSI) feedback scheme for Frequency Division Duplex (FDD) massive multiple-input multipleoutput (MIMO) systems in U6G band (6425MHz-7125MHz). Existing compressive sensing (CS)-based and deep learning (DL)- based schemes preprocess the channel by truncating it in the angular-delay domain. However, the energy leakage effect caused by the Discrete Fourier Transform (DFT) basis will be more serious and leads to a bottleneck in recovery accuracy when applied to wideband channels such as those in U6G. To solve this problem, we introduce the Loewner Interpolation (LI) framework which generates a set of dynamic bases based on the current CSI matrix, enabling highly efficient compression in the frequency domain. Then, the LI basis is further compressed in the spatial domain through a neural network. To achieve a flexible trade-off between feedback overhead and recovery accuracy, we design a rateless auto-encoder trained with tail dropout and a multi-objective learning schedule, supporting variable-length feedback with a singular model. Meanwhile, the codewords are ranked by importance, ensuring that the base station (BS) can still maintain acceptable reconstruction performance under limited feedback with tail erasures. Furthermore, an adaptive quantization strategy is developed for the feedback framework to enhance robustness. Simulation results demonstrate that the proposed scheme could achieve higher CSI feedback accuracy with less or equal feedback overhead, and improve spectral efficiency compared with baseline schemes.

Figures (9)

• Figure 1: The LI-based CSI interpolation model.
• Figure 2: The illustration of the LI-based CSI compression scheme.
• Figure 3: The illustration of the proposed LI-MORnet.
• Figure 4: The workflow of the overall framework.
• Figure 5: The Training Curve of the LI-MORNet.