Knowledge and Data Dual-Driven Channel Estimation and Feedback for Ultra-Massive MIMO Systems under Hybrid Field Beam Squint Effect

TL;DR

Simulation results show that the proposed knowledge and data dual-driven approach outperforms conventional downlink CE and CSI feedback methods, especially in the case of low signal-to-noise ratios.

Abstract

Acquiring accurate channel state information (CSI) at an access point (AP) is challenging for wideband millimeter wave (mmWave) ultra-massive multiple-input and multiple-output (UMMIMO) systems, due to the high-dimensional channel matrices, hybrid near- and far- field channel feature, beam squint effects, and imperfect hardware constraints, such as low-resolution analog-to-digital converters, and in-phase and quadrature imbalance. To overcome these challenges, this paper proposes an efficient downlink channel estimation (CE) and CSI feedback approach based on knowledge and data dual-driven deep learning (DL) networks. Specifically, we first propose a data-driven residual neural network de-quantizer (ResNet-DQ) to pre-process the received pilot signals at user equipment (UEs), where the noise and distortion brought by imperfect hardware can be mitigated. A knowledge-driven generalized multiple measurement vector learned approximate message passing (GMMV-LAMP) network is then developed to jointly estimate the channels by exploiting the approximately same physical angle shared by different subcarriers. In particular, two wideband redundant dictionaries (WRDs) are proposed such that the measurement matrices of the GMMV-LAMP network can accommodate the far-field and near-field beam squint effect, respectively. Finally, we propose an encoder at the UEs and a decoder at the AP by a data-driven CSI residual network (CSI-ResNet) to compress the CSI matrix into a low-dimensional quantized bit vector for feedback, thereby reducing the feedback overhead substantially. Simulation results show that the proposed knowledge and data dual-driven approach outperforms conventional downlink CE and CSI feedback methods, especially in the case of low signal-to-noise ratios.

Figures (22)

• Figure 1: Structure of the knowledge and data dual-driven network: the whole network is divided into data-driven de-quantization, knowledge-driven GMMV-LAMP, and data-driven CSI feedback modules. Time-domain quantized oversampling samples are first inputted into the de-quantization module, and the output is utilized in GMMV-LAMP for CE. Finally, the estimated CSI is compressed and reconstructed in the CSI feedback module.
• Figure 2: Illustration of the de-quantization: the top subfigure depicts the analog perfect received signal without noise or distortion and the analog distorted received signal; the middle subfigure depicts the sampled distorted received signal and its quantized version under 4-time oversampling; the bottom subfigure compares the analog perfect received signal with the sampled signal after ResNet-DQ, which demonstrates that most of the information has been recovered.
• Figure 3: The transmission region can be split into far-field and near-field regions.
• Figure 4: Far-field beam squint effect for the case of 2 multipath components.
• Figure 5: Near-field beam squint effect for the case of 2 multipath components.