Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enabling Feedback-Free MIMO Transmission for FD-RAN: A Data-driven Approach

Jingbo Liu, Jiacheng Chen, Zongxi Liu, Haibo Zhou

TL;DR

This work addresses the challenge of MIMO transmission in fully-decoupled FD-RAN where CSI feedback is infeasible. It proposes two data-driven, end-to-end MIMO solutions that map UE geolocation to transmission parameters, circumventing CSI feedback: a codebook-based approach using 5G CLSM principles and a non-codebook SVD-based approach that leverages a Variational Autoencoder to compress precoders and Gaussian Process Regression for space-domain inference. The SVD-based method, combined with VAE latent representations and GPR-based location interpolation, shows superior spatial inference and reduces zero-throughput locations relative to codebook-based methods, achieving competitive performance with significantly reduced feedback overhead. Extensive simulations on 5G link-level and DeepMIMO ray-tracing data demonstrate the potential of data-driven MIMO for FD-RAN and future 6G deployments, while highlighting the importance of conservative RI decisions and accurate latent-variable forecasting. The results suggest that end-to-end, feedback-free MIMO can approach the performance of CLSM in many scenarios, with meaningful gains in overhead reduction and adaptability to dynamic environments.

Abstract

To enhance flexibility and facilitate resource cooperation, a novel fully-decoupled radio access network (FD-RAN) architecture is proposed for 6G. However, the decoupling of uplink (UL) and downlink (DL) in FD-RAN makes the existing feedback mechanism ineffective. To this end, we propose an end-to-end data-driven MIMO solution without the conventional channel feedback procedure. Data-driven MIMO can alleviate the drawbacks of feedback including overheads and delay, and can provide customized precoding design for different BSs based on their historical channel data. It essentially learns a mapping from geolocation to MIMO transmission parameters. We first present a codebook-based approach, which selects transmission parameters from the statistics of discrete channel state information (CSI) values and utilizes integer interpolation for spatial inference. We further present a non-codebook-based approach, which 1) derives the optimal precoder from the singular value decomposition (SVD) of the channel; 2) utilizes variational autoencoder (VAE) to select the representative precoder from the latent Gaussian representations; and 3) exploits Gaussian process regression (GPR) to predict unknown precoders in the space domain. Extensive simulations are performed on a link-level 5G simulator using realistic ray-tracing channel data. The results demonstrate the effectiveness of data-driven MIMO, showcasing its potential for application in FD-RAN and 6G.

Enabling Feedback-Free MIMO Transmission for FD-RAN: A Data-driven Approach

TL;DR

This work addresses the challenge of MIMO transmission in fully-decoupled FD-RAN where CSI feedback is infeasible. It proposes two data-driven, end-to-end MIMO solutions that map UE geolocation to transmission parameters, circumventing CSI feedback: a codebook-based approach using 5G CLSM principles and a non-codebook SVD-based approach that leverages a Variational Autoencoder to compress precoders and Gaussian Process Regression for space-domain inference. The SVD-based method, combined with VAE latent representations and GPR-based location interpolation, shows superior spatial inference and reduces zero-throughput locations relative to codebook-based methods, achieving competitive performance with significantly reduced feedback overhead. Extensive simulations on 5G link-level and DeepMIMO ray-tracing data demonstrate the potential of data-driven MIMO for FD-RAN and future 6G deployments, while highlighting the importance of conservative RI decisions and accurate latent-variable forecasting. The results suggest that end-to-end, feedback-free MIMO can approach the performance of CLSM in many scenarios, with meaningful gains in overhead reduction and adaptability to dynamic environments.

Abstract

To enhance flexibility and facilitate resource cooperation, a novel fully-decoupled radio access network (FD-RAN) architecture is proposed for 6G. However, the decoupling of uplink (UL) and downlink (DL) in FD-RAN makes the existing feedback mechanism ineffective. To this end, we propose an end-to-end data-driven MIMO solution without the conventional channel feedback procedure. Data-driven MIMO can alleviate the drawbacks of feedback including overheads and delay, and can provide customized precoding design for different BSs based on their historical channel data. It essentially learns a mapping from geolocation to MIMO transmission parameters. We first present a codebook-based approach, which selects transmission parameters from the statistics of discrete channel state information (CSI) values and utilizes integer interpolation for spatial inference. We further present a non-codebook-based approach, which 1) derives the optimal precoder from the singular value decomposition (SVD) of the channel; 2) utilizes variational autoencoder (VAE) to select the representative precoder from the latent Gaussian representations; and 3) exploits Gaussian process regression (GPR) to predict unknown precoders in the space domain. Extensive simulations are performed on a link-level 5G simulator using realistic ray-tracing channel data. The results demonstrate the effectiveness of data-driven MIMO, showcasing its potential for application in FD-RAN and 6G.
Paper Structure (22 sections, 21 equations, 16 figures)

This paper contains 22 sections, 21 equations, 16 figures.

Table of Contents

  1. Introduction
  2. Related Works
  3. System Model and Problem Formulation
  4. System Model
  5. Problem Formulation
  6. Codebook-based Approach
  7. Codebook-based 5G CLSM
  8. Statistic-based Solution
  9. SVD-based Approach
  10. VAE-based Solution
  11. Fixing of Transmission Parameters
  12. Inference of Transmission Parameters
  13. Simulation Results and Discussions
  14. Simulation Setup
  15. Environment
  16. ...and 7 more sections

Figures (16)

  • Figure 1: System model.
  • Figure 2: Flow charts of proposed approaches for data-driven MIMO.
  • Figure 3: Our proposed VAE-based solution to dealing with optimal precoders of rank 4 obtained from SVD of channels.
  • Figure 4: Locations of selected UEs and BS in DeepMIMO O1 scenario in the $x$ and $y$ directions.
  • Figure 5: Average throughput comparison between CLSM and statistic-based solutions when transmitting single and multiple streams.
  • ...and 11 more figures