A Wideband CSI Feedback Scheme Based on Loewner Interpolation and Model Order Reduction

In Frequency Division Duplex (FDD) massive multiple-input multiple-output (MIMO) systems, it is vital for the base station (BS) to obtain accurate downlink channel state information (CSI). In 2022, 3GPP officially authorized the U6G spectrum, marking this band as a potential spectrum hotspot in the next generation communication systems. Under such huge bandwidth, ensuring high-precision CSI feedback while controlling the feedback overhead becomes a key technical challenge. In this paper, we propose a novel wideband CSI feedback scheme based on Loewner interpolation (LI) and model order reduction (MOR), which achieves highly efficient compression in both spatial and frequency domains. Instead of relying on a fixed basis (such as DFT basis), the LI-based CSI interpolation model is directly constructed according to the current CSI data sample, and thus is dynamically adjusted to guarantee high-precision recovery. Meanwhile, the MOR technique further prevents the expansion of feedback overhead of model parameters. Simulation results show that the proposed scheme could achieve higher CSI feedback precision with less or equal feedback overhead compared with the compressive sensing (CS) based method, deep learning (DL) based method and R16 Enhanced Type II (eType II) feedback scheme, especially for the wideband scenario.